ã - Bournemouth Universityeprints.bournemouth.ac.uk/31168/1/DE SILVA KANAKARATNE, Mahe… · ï z...

1

The Impact of National Culture and Industry Structure on Grocery Retail

Customer Loyalty: A Comparative Study of the United Kingdom and Sri

Lanka

Maheshan De Silva Kanakaratne

A thesis submitted in partial fulfilment of the requirements of Bournemouth University for the degree of Doctor of Philosophy

January 2018

Bournemouth University

2

Abstract

Despite the pivotal influence of customer loyalty on retailer performance and profitability, the

effect of national culture and industry structure on customer loyalty in grocery retailing has been

thinly researched. This PhD is the first to investigate the influence of these factors by examining

two culturally and structurally disparate markets; the UK and Sri Lanka. A sequential mixed

methods research design was used where a qualitative phase (three focus groups in each country)

preceded a large scale survey (n= 550). The preliminary inductive phase generated new knowledge

in the absence of adequate existing theory and guided the development of the survey instrument

which was used to further investigate the preliminary findings.

This study finds that national culture has an indirect influence on customer loyalty via customer

service and loyalty programmes. These findings make a significant contribution to theory by

challenging the limited existing literature that suggests a direct influence. A direct influence of

industry structure is identified in both countries; UK customers are more ‘truly’ loyal to their main

retailer and this is influenced by a common focus on quality by retailers. Similarly, greater spurious

and latent loyalty is evident in Sri Lanka where customers prefer retailers offering higher standards

of customer service and product quality. An indirect influence is also highlighted by the aversion to

tracking in the UK and openness in Sri Lanka which reflects the level of loyalty programme

maturity and customer familiarity in the two countries. These findings provide the first focused

insight of the influence of industry structure on customer loyalty.

Findings further highlight that national culture influences customer perceptions of loyalty

programme attributes where Sri Lankan customers are more open to tracking and tiered schemes

whilst expecting preferential treatment and soft benefits. Whilst supporting the limited existing

literature, these findings provide novel insights that will guide future research on the influence of

national culture on loyalty programmes. The theoretical framework provided in this study is the

first detailed depiction of customer loyalty in relation to national culture and industry structure.

This framework will act as a valuable theoretical base when future research is undertaken in this

area.

From a practical perspective, this research will enable domestic grocery retailers in Western

markets to understand how national culture influences customer loyalty, thereby, allowing them

to cater to culturally disparate customer segments more efficiently. Similarly, international

retailers can obtain a greater understanding of cultural and structural influences when developing

their strategies by focusing on key elements such as customer service, product quality and loyalty

programmes.

3

List of Contents

Chapter 1: Introduction ......................................................................................................................... 11

1.1 Research Rationale ...................................................................................................................... 12

1.2 National Culture .......................................................................................................................... 13

1.3 Industry Structure ....................................................................................................................... 14

1.4 Customer Loyalty ......................................................................................................................... 15

1.5 Loyalty programmes .................................................................................................................... 15

1.6 Study Focus ................................................................................................................................. 16

1.7 Background ................................................................................................................................. 17

1.7.1 United Kingdom .................................................................................................................... 17

1.7.2 Sri Lanka ............................................................................................................................... 18

1.8 Research Aim & Objectives .......................................................................................................... 18

1.9 Methodological Overview ............................................................................................................ 19

1.10 Thesis Structure ......................................................................................................................... 20

Chapter 2 .............................................................................................................................................. 22

Section 1: National Culture.................................................................................................................... 22

2.1 Introduction ................................................................................................................................ 23

2.2 Definitions of Culture .................................................................................................................. 23

2.3 Cultural Models ........................................................................................................................... 26

2.3.1 Hofstede’s Cultural Model .................................................................................................... 26

2.3.2 Trompenaars’ Cultural Model ............................................................................................... 27

2.3.3 The GLOBE Study .................................................................................................................. 28

2.3.4 Identifying Common Themes ................................................................................................ 29

2.3.5 Critique of Cultural Models ................................................................................................... 31

2.4 Section Summary ......................................................................................................................... 35

Section 2: Industry Structure ................................................................................................................. 36

2.5 Grocery Retailing in the United Kingdom ..................................................................................... 37

2.6 Grocery Retailing in Sri Lanka ...................................................................................................... 39

2.7 Comparison of the Grocery Retail Structures in the UK and Sri Lanka ........................................... 41

2.7.1 Enterprise Density ................................................................................................................ 41

2.7.2 Market Concentration........................................................................................................... 41

2.7.3 Product Sector Balance ......................................................................................................... 42

2.7.4 Employment Profile .............................................................................................................. 42

2.7.5 Innovation Levels .................................................................................................................. 42

4

2.8 Section Summary ......................................................................................................................... 43

2.9 Chapter Summary ........................................................................................................................ 43

Chapter 3: Customer Loyalty in Grocery Retailing .................................................................................. 44

3.1 Introduction ................................................................................................................................ 45

3.2 A Shift in Marketing Philosophy- Relationship Marketing ............................................................. 45

3.3 Customer Loyalty as a Theoretical Construct ............................................................................... 47

3.4 Defining Customer Loyalty ........................................................................................................... 49

3.5 Segmenting Customer Loyalty ..................................................................................................... 51

3.6 Customer Loyalty in Grocery Retailing ......................................................................................... 56

3.6.1 Measuring and Defining Customer Loyalty in Grocery Retailing ............................................. 56

3.6.2 Segmenting Customer Loyalty in Grocery Retailing ............................................................... 58

3.6.3 Antecedents of Customer Loyalty in Grocery Retailing .......................................................... 59

3.7 Loyalty Programmes in Grocery Retailing ..................................................................................... 64

3.7.1 Loyalty Programme Rewards ................................................................................................ 66

3.7.2 Concerns Related to Loyalty Programmes ............................................................................. 67

3.8 Customer Loyalty across Cultures ................................................................................................ 68

3.9 Cross-Cultural Retail Store Loyalty and Loyalty Programmes ........................................................ 70

3.10 Structural Impact on Customer Loyalty in Grocery Retailing ....................................................... 72

3.11 Literature Review Summary & Conceptual Framework Development ........................................ 73

Chapter 4: Methodology ....................................................................................................................... 75

4.1 Introduction ................................................................................................................................ 76

4.2 Research Philosophy .................................................................................................................... 76

4.3 Mixed Methods Research ............................................................................................................ 77

4.4 Empirical Study 1: Focus Group Discussions ................................................................................. 81

4.4.1 Focus Group Content and Discussion Guide .......................................................................... 82

4.4.2 Sampling ............................................................................................................................... 83

4.4.3 Ethical Considerations........................................................................................................... 85

4.4.4 Data Analysis ........................................................................................................................ 86

4.5 Empirical Study 2: Quantitative Surveys ....................................................................................... 87

4.5.1 Designing the Questionnaire ................................................................................................. 88

4.5.2 Questionnaire Content ......................................................................................................... 90

4.5.3 Pre-test and Translation ........................................................................................................ 92

4.5.4 Questionnaire Type .............................................................................................................. 93

4.5.5 Sampling Approach and Administration ................................................................................ 95

5

4.5.6 Data Analysis ...................................................................................................................... 101

4.6 Chapter Summary ...................................................................................................................... 110

Chapter 5 ............................................................................................................................................ 111

Section 1: Preliminary Interpretation of Qualitative Findings ............................................................... 111

5.1 Introduction .............................................................................................................................. 112

5.2 First Empirical Study- Focus Group Discussions .......................................................................... 112

5.2.1 Grocery Shopping Habits..................................................................................................... 112

5.2.2 Factors Affecting Store Choice ............................................................................................ 117

5.2.3 Differences in Customer Loyalty .......................................................................................... 121

5.2.4 Loyalty Programmes ........................................................................................................... 126

5.2.5 Section Summary & Refined Conceptual Framework Development ..................................... 129

Section 2: Quantitative Results ............................................................................................................ 131

5.3 Second Empirical Study- Questionnaires .................................................................................... 132

5.3.1 Sample Characteristics- UK ................................................................................................. 132

5.3.2 Sample Characteristics- Sri Lanka ........................................................................................ 133

5.3.3 Preliminary Composition of Items: Exploratory Factor Analysis ........................................... 134

5.3.4 Confirmatory Factor Analysis .............................................................................................. 135

5.3.5 Inter-Country Differences ................................................................................................... 136

5.3.6 Within-Country Differences: United Kingdom ..................................................................... 139

5.3.7 Within-Country Differences: Sri Lanka ................................................................................. 143

5.3.8 Summary and Confirmatory Framework Development ....................................................... 146

Chapter 6: Discussion ......................................................................................................................... 148

6.1 Introduction .................................................................................................................................. 149

6.2 Differences in Loyalty ................................................................................................................ 149

6.3 Overall Loyalty in the Two Countries .......................................................................................... 153

6.4 Loyalty Programmes .................................................................................................................. 155

6.4.1 Attitudes towards Tracking ................................................................................................. 155

6.4.2 Rewards .............................................................................................................................. 156

6.4.3 Soft Benefits, Preferential Treatment and Tiered Schemes .................................................. 158

6.5 Grocery Shopping Habits ........................................................................................................... 160

6.5.1 Aversion to Own Labels and Influence on Customer Loyalty ................................................ 160

6.5.2 Special Occasions ................................................................................................................ 162

6.5.3 Aversion to Fresh Produce .................................................................................................. 163

6.6 Importance of Customer Service ................................................................................................ 165

6

6.7 Chapter Summary and Theoretical Framework .......................................................................... 166

Chapter 7: Conclusion ......................................................................................................................... 169

7.1 Introduction .............................................................................................................................. 170

7.2 Revisiting the Research Aim & Objectives .................................................................................. 171

7.3 Summary of Key Findings ........................................................................................................... 173

7.4 Theoretical & Methodological Contributions ............................................................................. 175

7.5 Practical Implications & Recommendations to Grocery Retailers ............................................... 178

7.5.1 Differences in Loyalty.......................................................................................................... 178

7.5.2 Loyalty Programmes ........................................................................................................... 179

7.5.3 Own Labels ......................................................................................................................... 180

7.5.4 Convenience Stores ............................................................................................................ 181

7.6 Research Scope & Limitations .................................................................................................... 181

7.7 Recommendations for Future Research ..................................................................................... 182

Reference List ................................................................................................................................. 184

Appendix A: Focus Group Discussion Guide ..................................................................................... 221

Appendix B: Participant Information Sheet ...................................................................................... 225

Appendix C: Consent Form .............................................................................................................. 227

Appendix D: Questionnaire ............................................................................................................. 228

Appendix E: Store Image Items ........................................................................................................ 232

Appendix F: True Loyalty Scales ....................................................................................................... 232

Appendix G: Cover Letter ………………………………………………………………………………………. 230

Appendix H: Perceptual Map ........................................................................................................... 234

Appendix I: EFA Pattern Matrix- UK ................................................................................................. 235

Appendix J: EFA Pattern Matrix- Sri Lanka ....................................................................................... 236

Appendix K: Composite Reliability, Converegent Validity & Divergent Valdity- UK ........................... 237

Appendix L: Composite Reliability, Converegent Validity & Divergent Valdity- Sri Lanka .................. 237

7

List of Figures

Figure 2.1: Culture as an Onion 24

Figure 2.2: Comparison of the United Kingdom and Sri Lanka 34

Figure 2.3: Market Share Change 38

Figure 3.1: The Re-Emergence of Relationship Marketing 46

Figure 3.2: Relative Attitude-Repeat Patronage Loyalty Segments 51

Figure 3.3: Subdividing “Loyals” 52

Figure 3.4: Diamond of Loyalty 54

Figure 3.5: Oliver’s Four Stage Loyalty Model 55

Figure 3.6: Relative Attitude-Repeat Patronage Loyalty Segments 59

Figure 3.7: Key Antecedents of Customer Store Loyalty 63

Figure 3.8: Conceptual Framework 74

Figure 4.1: Main Stages in the Research Process 79

Figure 4.2: Research Design 80

Figure 4.3: Survey Coverage of the United Kingdom 96

Figure 4.4: Survey Coverage of Sri Lanka 99

Figure 5.1: Key Themes and Sub Themes Pertaining to Grocery Shopping Habits 112

Figure 5.2: Key Findings pertaining to Grocery Shopping Habits 117

Figure 5.3: Key Attributes of Store Image Influencing Customer Loyalty 121

Figure 5.4: Key Findings pertaining to Differences in Customer Loyalty 125

Figure 5.5: Key Findings pertaining to Loyalty Programmes 129

Figure 5.6: Refined Conceptual Framework 130

Figure 5.7: Confirmatory Framework 147

Figure 6.1: Loyalty to Main Grocery Retailer in the UK 149

Figure 6.2: Loyalty to Main Grocery Retailer in Sri Lanka 151

Figure 6.3: Influence of national culture and industry structure on customer loyalty 155 in the UK and Sri Lanka

Figure 6.4: The influence of national culture and industry structure on customer 160 perceptions of loyalty programme attributes

Figure 6.5: The influence of industry structure on customer loyalty through grocery 165 shopping habits

Figure 6.6: Theoretical Framework 168

8

List of Tables

Table 2.1: GLOBE versus Hofstede 33

Table 2.2: Channel Performance in UK Grocery Retailing 37

Table 2.3: Market Shares of Sri Lankan Organised Grocery Retailers 39

Table 2.4: Structural Differences between the UK and Sri Lanka 43

Table 3.1: Benefits of Customer Loyalty 47

Table 3.2: Advantages of Attitudinal Loyalty 48

Table 3.3: Attributes of Retail Mix or Store Image 60

Table 3.4: Studies on Cross-Cultural Customer Loyalty using Hofstede’s Dimensions 69

Table 4.1: Characteristics of Participants at the time of data collection (April to August 2015) 84

Table 4.2: Response Rates 101

Table 5.1: The Top Five Factors Affecting Store Choice 117

Table 5.2: Characteristics of the UK sample 132

Table 5.3: Characteristics of the Sri Lankan sample 134

Table 5.4: Model Fit for UK and Sri Lanka 135

Table 5.5: Differences between the UK and Sri Lanka 137

Table 5.6: Degree of importance placed on attributes of Store Image 138

Table 5.7: Differences in customer loyalty towards the main supermarket 140

Table 5.8: Supermarket availability and customer loyalty 140

Table 5.9: Supermarket usage and customer loyalty 141

Table 5.10: Reasons for Monthly Supermarket Usage 141

Table 5.11: Differences in customer loyalty towards the main supermarket 143

Table 5.12: Supermarket availability and customer loyalty 144

Table 5.13: Reasons for Monthly Supermarket Usage 144

Table 6.1: Research Findings versus Previous Findings 154

9

Acknowledgment

Whilst being a challenging experience, this PhD has enabled me to conduct research in an area

that has interested me, even as a child. I am grateful to many people for their support during this

journey but the following cannot go unmentioned.

My sincere gratitude goes to my first supervisor, Doctor Jeffery Bray for his enthusiasm, support

and honest feedback. I am immensely grateful for his understanding during the various occasions

where personal matters hampered my morale. I am also grateful to Associate Professor Julie

Robson for her supervisory guidance during the final six months of this PhD; thank you for the

valuable advice and feedback on my work. I would also like to thank Doctor Charles McIntyre who

supervised this PhD for nearly three years. I am thankful to Doctor John Beavis for his advice on

my statistical analyses.

I am thankful to Thivanka, my wife whose love and support were crucial. I am amazed by your

courage and am humbled by your many sacrifices to help me pursue this PhD. To our son,

Tishayne who is unaware of why his father is at times grumpy; you have lit up my world. I am

indebted to my parents; Ammi and Thaththi, thank you for your love and support. I am also

grateful to Thivanka’s parents for their support, especially during this journey. A special thanks to

Sheran and Nilushi for being there for Thivanka and Tishayne when I was away from Sri Lanka. I am

grateful to Anura Uncle, Bukka and Julius Uncle for their kindness and support when I needed

them the most.

Finally, I am grateful to Bournemouth University for offering me a scholarship and the many

opportunities to grow as an academic.

10

Declaration

This research has been presented/will be presented at the following conferences:

De Silva Kanakaratne, M., 2015. The Impact of cultural and structural differences on

grocery retail consumer loyalty and their implications on grocery retail loyalty programme

design: A cross-cultural study of the United Kingdom and Sri Lanka. 1st Academy of

Marketing Consumer Research SIG PhD Colloquium. 27 April, Chester UK.

De Silva Kanakaratne, M., 2016. The Impact of cultural and structural differences on

grocery retail consumer loyalty and their implications on grocery retail loyalty programme

design: A cross-cultural study of the United Kingdom and Sri Lanka. The 8th Bournemouth

University Annual Postgraduate Research Conference. 9-10 March, Bournemouth UK.

De Silva Kanakaratne, M., Bray, J. and Robson, J., 2018. Exploring the Influence of National

Culture and Industry Structure on Grocery Retail Customer Loyalty. Colloquium on

European Research in Retailing. 11-13 July, Guildford UK.

The following journal article has been published from a separate collaborative study:

Birch, D., Memery, J. and De Silva Kanakaratne, M., 2018. The mindful consumer:

Balancing egoistic and altruistic motivations to purchase local food. Journal of Retailing

and Consumer Services [online], 40 (January 2018), 221-228.

11

Chapter 1

Introduction

12

1.1 Research Rationale

Grocery retailing in the major Western European countries of the United Kingdom, Germany and

France has been experiencing high levels of competition and industry saturation (Assaf et al. 2012;

Eckman et al. 2015). The UK grocery retail sector in particular has been experiencing high levels of

competition and saturation since the 1980s (Duke 1989; 1992). Competition in the sector has been

exacerbated more recently by the entry and growth of deep discounters such as Aldi and Lidl.

Furthermore, the sector has witnessed consolidation with international and national retail chains

securing large portions of the market at the expense of smaller independent and traditional

retailers. Consolidation efforts have resulted in the four largest retailers (Tesco, Sainsbury’s, Asda

and Morrisons) also known as the Big Four, controlling over 70% of the market (Kantar Worldpanel

2018). ‘Push’ factors such domestic market saturation and increased competition (Poole et al.

2006; Alexander and Doherty 2009) as well as ‘Pull’ factors (Alexander 1990; Williams 1992) have

resulted in UK grocery retailers venturing into less developed international markets.

Despite the active international expansion habits of grocery retailers (Burt et al. 2008), the

influence of national culture on customer loyalty has been scarcely researched. This oversight by

international retailers has been identified in previous research (Straughan and Albers-Miller 2001;

Khare et al. 2014). Etgar and Rachman-Moore (2008) contend that failure and decline in retail

sales have been attributed to a lack of understanding of cultural differences in new markets. The

need to study the influence of national culture is further heightened by the growth of human

migration into and within Western Europe. This is due to the need to understand the diverse

needs of different customer groups as a result of domestic markets becoming increasingly multi-

cultural. Despite limited research in this area, it is clear that national culture plays a critical role in

influencing heterogeneous consumer behaviour (Seock and Lin 2011). As a result, retailers aiming

to enter foreign markets or those who cater to immigrants and ethnic customers in domestic

markets need to examine the impact of cultural differences when formulating their strategies

(Akehurst and Alexander 1995; Thomas and Carraher 2014).

Whilst different structural conditions can be expected in new markets (Laaksonen 1993), previous

research has not adequately focused on the impact of such differences on customer loyalty. This

lack of previous research exists regardless of contentions that the performance of expanding

retailers is dependent on a market’s structural factors (Chan et al. 2011). In a similar vein, culture’s

influence on customer perceptions and behaviour towards loyalty programmes remains thinly

explored (Park et al. 2013; Hu and Weber 2014; Beck et al. 2015). This is rather surprising given

13

the pervasiveness of such programmes in grocery retailing, particularly in developed markets.

Given the growth of internationalisation and the diversification of domestic customers, this study

was designed to investigate the impact of national culture and industry structure on customer

loyalty in the context of grocery retailing.

1.2 National Culture

The word ‘Culture’ has been used in various contexts, providing various meanings to people.

Consequently, the word is considered to be one of the most difficult and complex words in the

English language (Williams 1976; de Mooij 2010). For the purpose of this study, an anthropological

view of the word is used as it provides an appropriate definition; relating to how humans behave.

Thus, culture could be described as a collection of ideas, beliefs, values, knowledge, behaviour,

customs and norms shared by a country, society or even a group of people that could be passed

over from one generation to another (Collins English Dictionary 2018; Oxford English Dictionary

2018). This concept, when used to study people in different nations is known as National Culture.

National culture can be described as a set of enduring personality traits common to people in a

country (Clark 1990). These traits define ways of thinking, feeling and acting and are influenced by

common values and conventions of a society (Nakata and Sivakumar 2001). These common values

could be influenced by similar backgrounds, education and life experiences (Doney et al. 1998).

Whilst differences do exist between countries, understanding national culture has proven to be a

crucial, yet difficult task (Clark 1990). Research interest in national culture has persisted however

since the 1950s due to the identification that customer perceptions, behaviour and decision

making are all influenced by differences in national culture (Clark 1990; Straughan and Albers-

Miller 2001).

Over the years, several models of national culture have been proposed by various authors. Of

those, Hofstede’s (1980) dimensions of national culture with subsequent additions; Hofstede and

Bond (1988) and Hofstede et al. (2010) remain the most widely cited dimensions in cross-cultural

consumer behaviour studies. Whilst widely applied in marketing studies (Nakata and Sivakumar

2001; Laroche et al. 2004), this model has also received criticism from various scholars. However,

this model was used to provide a theoretical base for this study, supported by the following

reasons.

14

I. Its widespread use in the existing literature provided a reference point for the research

findings to be compared against

II. This model being the only model applied to consumer behaviour, loyalty and loyalty

programmes enabled a preliminary understanding of culture’s influence on these aspects

III. This model being the only one with dimension scores for the UK and Sri Lanka provided a

basis for comparison which was crucial to identify differences and similarities

1.3 Industry Structure

Industry structure in the context of this study is viewed as the current nature of competitive rivalry

amongst grocery retailers (Akehurst and Alexander 1995). Understanding the nature of domestic

as well as foreign markets is important as conditions may not always be the same in different

markets (Laaksonen 1993). Studying the structure of a market provides a good indication of its

level of development and unique characteristics within that market; particularly to retailers

entering new markets (Alexander and Doherty 2009). Therefore, success or failure for a retailer

would be influenced by how well these various structural factors are identified and understood

(Chan et al. 2011).

Disparate structural development conditions are evident in the UK and Sri Lanka. Whilst the

organised sector constitutes nearly 94% of grocery retailing in the UK (IGD Retail Analysis 2017b),

the predominant traditional sector makes up approximately 84% of grocery retailing in Sri Lanka

(LMRB 2016). Similarly, more than twenty organised grocery retail chains operate in the UK;

operating over 57,000 stores (IGD Retail Analysis 2017b). On the contrary, five retail chains

currently operate in Sri Lanka; with less than 800 stores. Own labels enjoy nearly 60% market

share in the UK with high levels of customer satisfaction (Key Note 2015). However, low levels of

own label penetration are evident in Sri Lanka. Online grocery retailing is a growing channel in the

UK (IGD Retail Analysis 2017b) but this is non-existent in Sri Lanka (LMD 2017). Finally, loyalty

programmes are widely used by UK grocery retailers and are used to track customer behaviour

and communicate with customers. However, Sri Lankan grocery retailers use such schemes merely

to offer customers discounts and offers.

15

1.4 Customer Loyalty

Customer loyalty has long been identified as a crucial requirement for business survival and

growth (Reichheld et al. 2000; Nguyen and Mutum 2012), even in the context of grocery retailing.

Despite being treated as a purely behavioural phenomenon in seminal research, customer loyalty

is best understood as a dichotomous construct, comprised of both behavioural and attitudinal

elements. This composite view of customer loyalty is argued to be more comprehensive, allowing

researchers to not only identify patterns of purchasing behaviour, but also the implicit attitudinal

factors that drive such behaviour (Day 1969; Dick and Basu 1994; Uncles et al. 2003).

Similarly, patronage has also been treated by various scholars in a purely behavioural manner. This

is despite early contentions (Howell and Rogers 1981), that purely behavioural measures alone

should not be applied in the study of retail store patronage. However, even recent research (Luceri

and Latusi 2012; Maruyama and Wu 2014) has treated patronage in a purely behavioural sense.

Other scholars (Grewal et al. 2003; Teller et al. 2013) as well as more recent research (Blut et al.

2018) have argued for the inclusion of attitudinal elements such as intentions, preferences and

satisfaction when measuring retail store patronage. Although it is clear that both customer loyalty

and retail store patronage should be measured as composite (behavioural and attitudinal)

measures, this study uses these two constructs differently to ensure clarity. Therefore, Customer

Loyalty henceforth will be used to denote attitudinal and behavioural loyalty whilst Patronage will

refer to purely behavioural (repeat) patronage behaviour.

1.5 Loyalty programmes

A loyalty programme is a collection of marketing initiatives aimed at improving customer loyalty

(Sharp and Sharp 1997; Leenheer et al. 2007). These tools are used by retailers to offer benefits to

members to create, enhance and maintain customer loyalty (Palmer et al. 2000; Yi and Jeon 2003;

Garcia-Gomez et al. 2006). These programmes are built on the premise of building personal

relationships with customers where dyadic communication is a key focus (Meyer-Waarden and

Benavent 2009). The adoption of loyalty programmes has been driven by the need to enhance

knowledge about customers (Mauri 2003). Loyalty schemes are also used to increase repeat

purchasing and obtain a greater proportion of spending from customers (Meyer-Waarden and

Benavent 2009).

16

Loyalty programmes are advanced in the UK with grocery retailers such as Tesco and Sainsbury’s

using them to gather customer data, track customer behaviour and communicate with individual

customers. On the contrary, such programmes are relatively infantile in Sri Lanka as grocery

retailers do not use customer data for tracking and individual communication; highlighting clear

structural differences. Whilst previous research on the impact of national culture on loyalty

programmes is scarce, the influence of industry structure is unknown. Therefore the UK and Sri

Lanka provide a firm platform for these influences to be explored.

1.6 Study Focus

The United Kingdom and Sri Lanka were chosen for this study due to their cultural and structural

differences. Whilst clearly distinguishable on several grounds of national culture, the two

countries also demonstrate acutely contrasting domestic retail structures. The United Kingdom

has a highly advanced grocery retail structure and has spawned several successful international

retailers. Considered to be a country in the first wave of modern retailing (Reardon et al. 2012),

the UK can be viewed as a ‘Global Market of Origin’ (Alexander and Doherty 2009). In light of the

United Kingdom’s decision to leave the European Union, South Asia has been identified as a good

match for UK retailer expansions due its potential; driven by economic growth in the region and

increasing customer spending power (Financial Times 2017). Sri Lanka can be used as a case study

by other Western European retailers aiming to expand into this territory. Furthermore, findings

from this study can also be used by grocery retail chains such as Carrefour who are currently

engaged in international markets.

Sri Lanka by contrast exhibits a developing structure and is considered to be a country in the third

wave of modern retailing (Reardon et al. 2012). The recent growth of modern retailing has made

Sri Lanka a country of interest to international retailers as a ‘Regional Market of Destination’

(Alexander and Doherty 2009). The island nation can also be viewed as a ‘gateway market’ in the

South Asian region due to its cultural and structural similarities with India and Pakistan (Pandey et

al. 2015). Successful expansion into this market can be used as a precursor to development in the

large Indian subcontinent; making Sri Lanka a market of particular interest to international

retailers. Given the cultural and structural similarities, findings from this study may be useful to UK

retailers such as Tesco and Sainsbury’s who already operate in India and Pakistan (IGD Retail

Analysis 2017c; The Express Tribune 2017). Sri Lanka is also attracting interest from global retailers

17

due to its steady economic development, rising per capita GDP and favourable government

policies towards foreign investments (Euromonitor International 2014; AT Kearney 2016).

Besides these differences, the choice of the UK was also due to the author being based in the UK

during the course of the study. Sri Lanka being his home country was identified as a suitable choice

due to reasons such as ease of access and local knowledge which proved to be invaluable when

collecting primary data. Access to Sri Lankan participants and respondents ensured the quality of

the data collected. Furthermore, cultural knowledge allowed the researcher to conduct the

research accordingly.

1.7 Background 1.7.1 United Kingdom

The annual sales value of the UK retail industry is approximately £388 billion (Office for National

Statistics 2017a); highlighting the magnitude of the industry. Food retailing constitutes almost 50%

of the market at £184.5 billion per year (IGD Retail Analysis 2017a). 93.6% of the sector is

comprised of modern retail formats; hypermarkets, supermarkets, convenience stores,

discounters and online. Further growth of these modern formats is expected with traditional

trading formats such as independent stores and specialist stores forecasted to grow only by 3.6%

by 2022 (IGD Retail Analysis 2017b). Furthermore, significant growth is expected in the

convenience, discount and online formats; signalling a possible change in the landscape of grocery

retailing.

Whilst over twenty grocery retailer chains operate in the UK, the Big Four; Tesco, Sainsbury’s, Asda

and Morrisons control over 70% market share (Kantar Worldpanel 2018). However, their

dominance has been checked, especially by the growth of Aldi and Lidl who have captured a

collective market share of nearly 12% (Kantar Worldpanel 2018). The growth of these two

discounters is driving the overall growth of the discount format; capturing market share from the

Big Four as well as other established retailers such as Co-op and Waitrose. Their Everyday-Low-

Price (EDLP) approach has resulted in the Big Four retailers also focusing more on lower prices

instead of promotions. This overall deep-discount ethos also appears to have influenced customer

expectations with the majority of customers expecting EDLPs (Mintel 2016a).

18

1.7.2 Sri Lanka

The Sri Lankan retail sector was valued at approximately £23.8 billion as at 2016 (LMD 2016), a

mere 5.8% of the value in the UK. However, the value of the grocery sector is unknown, perhaps

due to the predominantly unorganised nature of the sector. In stark contrast to the UK, grocery

retailing in Sri Lanka is dominated by so called unorganised retailing; independent grocery

retailers, independent grocery wholesalers and wet markets. This sector accounts for

approximately 84% of the entire grocery retailing sector’s value with over 240,000 stores in

operation (LMRB 2016). Despite the dominance of this sector, modern retailing formats have been

enjoying consistent growth of nearly 5% year-on-year (LMD 2016). Five grocery retail chains are

prominent in the country but one retailer; Cargills enjoys more than 40% of the organised market

(LMRB 2016). Overall, these five retailers control nearly 98% of the organised sector in value terms

with the rest being shared amongst various standalone grocery retailers. Compared to over 57,000

organised grocery retail stores found in the UK (IGD Retail Analysis 2017b), the five grocery

retailers operate less than 800 stores throughout the country. Whilst the Big Four retailers operate

throughout the UK, the market leader; Cargills is the only retailer with a nationwide coverage of

the country (LMD 2017). Modern grocery retailing is largely confined to supermarket type stores

while superstores or hypermarkets are not found in the country. Online grocery retailing is non-

existent (LMD 2017) and could be a result of the relative immaturity of organised grocery retailing

in the country.

1.8 Research Aim & Objectives

Despite increasing levels of retail internationalisation and the clear influence of national culture as

well as industry structure on various aspects of consumer behaviour, their influence on customer

loyalty remains thinly explored. Therefore, the aim and objectives listed below have been

developed to address the following research question.

Do national culture and industry structure influence customer loyalty in grocery retailing and if so,

what is the nature of their influence?

Research Aim

The aim of this research is to critically evaluate the influence of national culture and industry

structure on grocery retail customer loyalty in the United Kingdom and Sri Lanka.

19

Research Objectives

1. Develop a conceptual framework of grocery retail customer loyalty, national culture and

industry structure

2. Refine the conceptual framework by exploring the impact of national culture and industry

structure on grocery retail customer loyalty in the UK and Sri Lanka

3. Develop a confirmatory framework by further investigating the impact of national culture

and industry structure on grocery retail customer loyalty in the two countries

4. Present a theoretical framework highlighting the impact of national culture and industry

structure on grocery retail customer loyalty by synthesising the primary and secondary

data

5. Make recommendations to grocery retailers on how to enhance customer loyalty in the

UK and Sri Lanka

1.9 Methodological Overview

This study commenced with a thorough review of the existing literature on national culture and

customer loyalty. Industry data on the two countries were also reviewed with the pertinent

elements used to highlight disparities between the two countries. Primary data was collected

using a sequential mixed methods approach where findings from an inductive qualitative phase

informed the design of the subsequent quantitative survey. Three consumer focus groups were

conducted in each country. These discussions identified differences in grocery shopping behaviour,

attitudes towards loyalty programmes as well as differences in customer loyalty between the two

countries. Qualitative findings were then used to develop a questionnaire to be used in the

ensuing quantitative phase. A combined sample of 550 was obtained from the two countries

which enabled statistical analyses to be carried out.

20

1.10 Thesis Structure

The remainder of this thesis is structured in the following manner;

The literature review is divided into two chapters; chapters 2 and 3. Chapter 2 is comprised of two

main sections; national culture and industry structure. The first section introduces the concept of

culture before reviewing the models of national culture presented by scholars. The review of these

models is used to identify common themes amongst the various dimensions subsumed under each

of these models. The models are then critiqued to identify the most suitable model to be applied

in this study. The strengths and limitations of the selected model are discussed before applying its

dimension scores to compare and contrast the two countries. The second section reviews the

industry structures in the two countries. The most pertinent elements are discussed with some

reference to theory, before using these to highlight structural disparities between the two

countries.

Chapter 3 critically reviews the existing knowledge of the concept of customer loyalty. The chapter

commences with an overview and evolution of the relationship marketing concept. Customer

loyalty is discussed as a theoretical construct before analysing several definitions of this concept.

Several customer loyalty models are reviewed before selecting a suitable model to be used as the

theoretical base for this study. Customer loyalty in grocery retailing is then discussed with

reference to definitions, segments and antecedents. Loyalty programmes are then discussed,

followed by a review of previous research focusing on the impact of national culture on customer

loyalty and loyalty programmes. The limited research on the influence of industry structure on

grocery store loyalty is finally reviewed followed by a summary of the entire literature review;

leading to the development of the conceptual framework.

The methodology employed in this study is discussed in Chapter 4. The mixed methods approach

and its individual components; focus groups and surveys are detailed. Data collection and analysis

techniques are also discussed. Chapter 5 presents the key qualitative findings which are

summarised to refine the conceptual framework; illustrating the possible impact (positive or

negative) of the findings on customer loyalty. The characteristics of survey respondents in the two

countries as well as descriptive statistics are presented. Finally, the results of the statistical tests

are presented and summarised in a confirmatory framework.

21

The primary data presented in the previous chapter are discussed in relation to national culture

and industry structure in Chapter 6; highlighting how differences and similarities between the two

countries may be influenced by these factors. The chapter concludes with a theoretical

framework; illustrating the key findings and their links to elements of national culture and industry

structure.

Chapter 7 concludes this thesis by highlighting the key findings and identifies the contributions to

theory. Recommendations are made to domestic and international grocery retailers based on the

findings. The scope and limitations of the study are discussed after which recommendations for

future research are made.

22

Chapter 2

Section 1

National Culture

23

2.1 Introduction

This chapter is divided into two sections with the first section reviewing the concept of national

culture. The second section reviews and compares the structure of grocery retailing in the United

Kingdom and Sri Lanka. The first section will begin with a brief introduction of culture’s etymology

followed by a critical review of several definitions and conceptualisations of culture in the context

of this study. Three well-known cultural models; Hofstede et al. (2010), Trompenaars (1993) and

the GLOBE Study (House et al. 2004) are reviewed; leading to the selection of the most suitable

model for this study. The selected model will be used to compare and contrast the cultural

dimensions of the United Kingdom and Sri Lanka. The second section of this chapter provides an

analysis of grocery retailing in these two countries. This overview identifies similarities and

differences between the grocery retail industry structures of the two countries based on market

data and industry structure theory.

2.2 Definitions of Culture

The etymological roots of the word culture date back to the twelfth century when the term was

used to describe the process of cultivation or specifically, the “tilling of the soil” (Hofstede 1991).

Since then, its meaning has evolved and fragmented (Oxford English Dictionary 2018); making it

one of the most difficult and complex words in the English language to understand (Williams 1976;

de Mooij 2010). Due to the various contexts in which it has been used, culture could be defined

differently based on the researcher’s theoretical perspective and the context in which it is applied

(Kaul 2007). In the context of this study, an anthropological view or definition of culture is

appropriate as it focuses on the way humans behave.

de Mooij (2010) defines culture as an adhesive that binds groups of people together. The absence

of cultural patterns and organised systems of significant symbols would make it difficult for people

to live together. She defines culture as (p.48):

“what defines a human community, its individuals, and social organizations”.

Hofstede (1991, p.21) defines culture as “the collective programming of the mind which

distinguishes the members of one human group from another”. Culture is a collective phenomenon

as it, at least to a certain extent, is shared by people who live or have lived in a social environment

in which such a culture was adopted (Hofstede et al. 2010). Culture is not inborn but is learnt and

derived from the social environment in which one is brought up.

24

Hofstede (1991) argues that culture consists of four layers; Symbols, Heroes, Rituals and Values.

Symbols are words, gestures, pictures or objects and these are only recognised by people who

share a particular culture. Heroes are people who could be alive, dead, real or imaginary; whose

characteristics are highly valued in a culture and these heroes could be role models to people

within that culture. Rituals are collective activities that are carried out for their own sake but are

considered essential by people in a culture. Greetings, ways of showing respect, social and

religious ceremonies are examples of rituals. While the first three (referred to as practices) are

visible to outsiders, values are not. Values are what make up the core of a culture and are general

propensities to prefer certain states of affairs over others. This layered structure of culture is

likened by Hofstede (1991) to an onion with several layers as depicted by Figure 2.1.

Figure 2.1: Culture as an Onion (Hofstede 1991) Similar to Hofstede (1991), Trompenaars (1993) compares the concept of national culture to an

onion consisting three layers. The core layer represents the embedded assumptions that

determine the way people cope with their environment. The middle layer contains the norms and

values that determine the nature of things (good, bad, right or wrong). These values and norms

are not easily visible and identified by observers. The outer layer contains all the elements that are

visible to the external world such as behaviour, rituals and language. To understand a culture, one

needs to unpeel this onion layer by layer (Trompenaars and Hampden-Turner 1997).

Similarities can be observed between the conceptualisations of national culture as presented by

Hofstede (1991) and Trompenaars (1993). The outer layer of this onion and three layers (rituals,

heroes and symbols) in Hofstede’s onion are both visible to outsiders. While values are the core

layer in Hofstede’s onion, norms and values make up the middle layer in Trompenaars’ onion with

embedded assumptions being the core. For Hofstede, values are implicit and are not visible to

Values Rituals

Heroes

Symbols

Practices

25

outsiders while values and norms could albeit not easily, be visible to outsiders according to

Trompenaars. It can be argued that outsiders could identify values and norms by paying close

attention to the more explicit aspects such as behaviour, rituals and language (outer layers). The

basic assumptions that govern such values and norms will be implicit and will not be visible to

outsiders.

According to Hall (1981), a group of people from one culture will have a particular way of living, a

pattern of behaviour, similar attitudes and material possessions. This broad viewpoint suggests

that a frame of reference and understanding of one another is formed by people within a

particular culture based on how they have been brought up. For Trompenaars and Hampden-

Turner (1997), culture can be viewed as a shared way in which a group of people understands and

interprets the world. This definition is based on the premise that one culture could be different

from another in the way in which people select solutions to problems.

Although these definitions do not appear to be similar at first glance, they identify and highlight a

common theme. It is that people from one culture have a common yet unique set of behaviour,

attitudes and beliefs which differentiates them from people in another culture. Not only does

culture set a group of people apart from other culturally different groups, it acts as a catalyst

helping people within a group to relate to one another, understand one another and coexist. A

person absorbs the culture of their group or community from the early stages of socialisation

which then continues (Ackerman and Tellis 2001; Luna and Gupta 2001). Culture will therefore

have a direct and significant impact on a person’s beliefs, attitudes, norms, thinking patterns and

values that shape their behaviour (Kroeber and Parsons 1958). The influence culture has on

human behaviour suggests that consumption behaviour could also be influenced by culture and

this is confirmed by previous research (Lam 2007; Seock and Lin 2011; Mazaheri et al. 2014).

The ensuing section will review the models developed by Hofstede et al. (2010), Trompenaars

(1993) and the GLOBE Study (House et al. 2004) in their attempts to study national culture.

National culture can be described as the patterns of enduring personality traits of people living in

a country (Clark 1990). National culture shapes ways of thinking, feeling and acting rooted in the

common values and conventions of a society (Nakata and Sivakumar 2001). As a result, it is

unlikely for customers in countries with different national cultures to behave in a similar way. This

is confirmed by the influence national culture has been found to have on perceptions, decision

making styles and consumer behaviour (Clark 1990; Straughan and Albers-Miller 2001).

26

2.3 Cultural Models

A cultural model can be used to identify similarities and differences between two or more cultures

or subcultures. This can be done by applying cultural variables that have been developed using

customer data (Hoft 1996). Several studies have been conducted to identify cultural dimensions

that can be used to study human and consumer behaviour. Each model offers a set of cultural

dimensions that can be used to identify cultural characteristics of a country. Some of these models

provide measures or numerical ratings for countries and these have been widely applied in cross-

cultural consumer behaviour research. It should be noted that other models do exist and these

were reviewed prior to selecting the three models mentioned above. For instance, Kluckhohn and

Strodtbeck’s (1961) model was precluded due to the lack of its application in prior consumer

behaviour research. The models by Hall (1981) and Schwartz (1999) were not considered as they

do not provide enough dimensions; two and three respectively. Given the comparative nature of

this study, it was important for the selected model to provide several dimensions to aid a

comprehensive comparative base. Consequently, the models developed by Hofstede et al. (2010),

Trompenaars (1993) and the GLOBE Study (House et al. 2004) were selected due to the higher

number of dimensions offered by each model as well as their application in prior consumer

behaviour research. Following a review of these models, they will be critiqued; leading to the

selection of the most appropriate model for this study.

2.3.1 Hofstede’s Cultural Model

Through his earlier work, Hofstede (1980) developed arguably the most widely used cultural

model to this day. This study was based on a multinational survey of IBM employees that explored

personal values related to work. The underlying assumption was that cultural differences can be

distinguished based on differences in emphasis placed on these values. A survey comprising 100

items was administered to 116,000 employees in 38 different occupations across national IBM

subsidiaries. The survey was conducted in 20 different languages and scores were calculated for 72

countries following further replications and extensions. The initial project was undertaken at two

points in time; first in 1968 and the second in 1972. Matched samples of survey respondents in

different local IBM subsidiaries were obtained to ensure similar demographic characteristics. In

doing so, Hofstede argues that responses clearly highlight cultural differences as the only

differentiating factor was national culture.

Four cultural dimensions were developed based on the initial survey results (Hofstede 1980).

27

Power Distance explains how members with less power in an institution accept the unequal

distribution of power. In other words, this is the distance between people with authority and

those who come under their control. Collectivism versus Individualism refers to the extent to

which a group’s interests take precedence over individual interests where the former is greater in

more Collectivist cultures. Stronger ties between individuals can be observed in Collectivist

cultures compared to Individualistic cultures. Masculinity versus Femininity emphasises the social

roles of men versus women. Masculine cultures are driven by the need for assertiveness,

competition and the need to control the environment. Feminine cultures by contrast reinforce

nurturance, concern for relationships and living in harmony with the environment. Uncertainty

Avoidance is the extent to which members of one culture feel threatened by ambiguous or

unknown scenarios. High uncertainty avoidance cultures are more risk averse and intolerant of

ambiguities.

Hofstede added two more dimensions based on his work with Michael Bond (Hofstede and Bond

1988) and Michael Minkov (Hofstede et al. 2010). Long Term Orientation (based on Confusion

Dynamism) is the focus on future reward oriented virtues where perseverance and thriftiness are

valued. Past and present oriented values are fostered in short term oriented cultures. Short Term

Oriented cultures focus on more immediate reward oriented virtues. Indulgence versus Restraint

sees the relatively free gratification of basic and natural human desires related to enjoying life and

having fun in indulgent cultures. Restraint is a conviction that such gratification needs to be

curbed and regulated by strict social norms.

2.3.2 Trompenaars’ Cultural Model

The way people solve problems was used as the underlying factor in the study (Trompenaars

1993). Data was gathered from approximately 15,000 managers in 30 organisations across 50

countries using a questionnaire consisting 16 questions. In doing so, Trompenaars developed

seven cultural dimensions with five dimensions focusing on relationships among people and two

other dimensions; one focusing on time orientation and the other on man’s relationship with

nature. These dimensions are explained below.

Universalism/Particularism refers to the extent to which rules are applied to people irrespective of

relationships. While Universalists apply these rules equally, Particularism means rules are applied

differently based on relationships amongst people and circumstances that warrant such

applications. Individualism/Collectivism explains the extent to which people consider themselves

28

to be individual or part of a group. Specific/Diffuse is concerned with the extent to which various

roles of people are compartmentalised or generalised. Specific value oriented people tend to have

a clear distinction between their personal and public lives while the opposite can be seen in

diffuse value oriented people.

Neutral/Affective is the degree to which the public expression of emotions is accepted. Neutral

societies suppress such public expressions while such behaviour is accepted in affective societies.

Achievement/Ascription looks at whether a person is respected because of their accomplishments

or because of inherited status. Sequential/Synchronic emphasises how people manage their

activities in relation to the time that is available. In sequential cultures, time is treated as a straight

line with a sequence of different events where there is a proper order. In synchronic cultures, time

is like a circle where different things can occur simultaneously. Inner-directed/Outer-directed

defines the relationship between humans and nature. While inner-directed people believe that

they can control the environment to achieve their goals, outer-directed people believe that they

should live in harmony with nature when striving to accomplish their goals.

2.3.3 The GLOBE Study

The GLOBE (Globalization Leadership and Organizational Behavior Effectiveness) study builds on

previous work by Hofstede (1980; 1991) and was carried out to understand how leadership

processes are influenced by cultural differences (Shi and Wang 2011). The study was conducted in

62 countries and nine cultural dimensions were identified (House et al. 2004). Six dimensions are

similar to those of Hofstede (Hofstede et al. 2010) while the other three are unique (House et al.

2004). This study stands out from the other models discussed here because it measures not just

values but also practices related to culture; resulting in eighteen cultural scores. This allows

researchers to identify disparities between what people from different cultures value and what,

they in fact practice.

Power Distance is concerned with the extent to which people expect the equal distribution of

power. Uncertainty Avoidance is the extent to which people follow social norms, rules and

procedures so that they can reduce the unpredictability of future events. Humane Orientation

looks at how much value is placed by people on fairness, altruism and generosity. Institutional

Collectivism is the extent to which the collective distribution of resources and collective action are

practiced in a culture. In-group Collectivism is concerned with how much pride, cohesiveness and

loyalty is expressed by individuals within their organisations and families. Assertiveness is the

29

degree of assertiveness, aggression and confrontation involved in relationships among people.

Gender Egalitarianism is where gender differences are minimised and both sexes are treated

equally. Future Orientation looks at the extent to which people are mindful of future

consequences when engaging in activities. Performance Orientation is the extent to which people

value and promote individual excellence and performance improvement.

2.3.4 Identifying Common Themes

These three models can be used by researchers as they serve two purposes. Firstly, each model

provides a set of conceptually sound and/or empirically tested cultural dimensions upon which

cross-cultural comparisons can be made (Nardon and Steers 2006). Secondly, some of these

models provide numeric scores that allow researchers to highlight differences and similarities

between and among cultures based on these dimensional scores. Five common themes across

these three models can be identified.

I. Relationship with the Environment:

Masculinity/Femininity (Hofstede 1980) and Inner-directed/Outer-directed (Trompenaars

1993) can be applied as they refer to the extent of control people attempt to have over the

environment. Three GLOBE (2004) dimensions; Assertiveness, Performance Orientation and

Humane Orientation are also included in this theme although these focus on social

relationships rather than the environment.

II. Social Organisation:

This theme is concerned with the degree of group integration in a society or organisation.

Individualism/Collectivism (Hofstede 1980; Trompenaars 1993; GLOBE 2004) can be applied as

they look at the degree of group integration.

III. Power Distribution:

The focus of this theme is on the degree of power distribution within a society. Power Distance

dimensions of Hofstede (1980) and GLOBE (House et al. 2004) are linked to this theme as well

as the Gender Egalitarianism dimension of the GLOBE study which looks at the extent to which

gender differences are minimised. The Achievement/Ascription dimension of Trompenaars

(1993) can be linked to this theme although this dimension does not directly relate to power

distance. Instead, the way status is attributed in a society is taken as the differentiating factor.

30

IV. Rule Orientation:

This theme refers to the role played by rules in the reduction of uncertainty in a society and

three dimensions are linked herein. Uncertainty Avoidance dimensions of Hofstede (1980) and

GLOBE (House et al. 2004) are linked to this theme. The former looks at how societies tolerate

uncertainty and use rules. The latter looks at how societies attempt to reduce uncertainty by

using rules and regulations. The Universalism/Particularism dimension of Trompenaars (1993)

is also applied as it refers to the extent to which rules are applied to people regardless of

relationships. However, this dimension does not directly refer to the avoidance of uncertainty

and ambiguity.

V. Time Orientation:

Trompenaars (1993) uses the dimension; Sequential/Synchronic to look at how people execute

tasks in relation to time. Long Term Orientation and Future Orientation developed by Hofstede

and Bond (1988) and GLOBE (House et al. 2004) respectively, refer to the extent to which

actions are guided by concerns about the future.

It is evident that despite being developed at different times and in different contexts, these three

cultural models share common themes under which similar dimensions can be grouped. It should

be noted however, that the presence of common themes does not imply that all subsumed

dimensions are conceptually similar. As per the review of these models above, it is clear that some

of these dimensions have been conceptualised and defined differently. For instance, it is clear that

some of Trompenaars’ dimensions are dissimilar to those of Hofstede and GLOBE. As a

consequence, dimensions that come under a common theme may produce different

interpretations, results or incongruent numeric scores. Therefore, it is important for a researcher

to select one cultural model to study cross cultural differences. The next section will critically

appraise the model selected for this research project.

31

2.3.5 Critique of Cultural Models

Despite offering seven dimensions, the model developed by Trompenaars (1993) cannot be easily

applied to cross-cultural consumer behaviour studies due to some overlapping dimensions. This

lack of mutual exclusivity may have led to this model not receiving much attention in empirical

research (Yeganeh et al. 2009).

Hofstede’s cultural model remains the most widely cited to this day in cross-cultural consumer

behaviour research (Yoo et al. 2011). This is confirmed by the number of citations, reviews,

replications and validations (Sondergaard 1994; Kirkman et al. 2006; Zhang et al. 2008);

highlighting the influence of his work. The suitability of applying this model to assess retail

consumption differences across countries has been noted by de Mooij and Hofstede (2002). The

applicability of this model in international retailing studies; (Alexander and Doherty 2009) provides

support for its application in other retailing studies as argued by Straughan and Albers-Miller

(2001). Despite this model being the most accepted in terms of culture theory among business and

marketing scholars (Nakata and Sivakumar 2001; Laroche et al. 2004), it has come under intense

scrutiny and has received criticism from various authors.

Given the study’s focus on work related values, its applicability to marketing related research has

been questioned (Yeganeh et al. 2009). Some authors (Seock and Lin 2011) have questioned

whether these dimensions can be applied in the 21st century as the first four dimensions were

developed 30 years ago. Kirkman et al. (2006) suggest that that these dimensions may have lost

their predictive validity over the years. However, other authors contend that the many replications

of the initial study in various research contexts in fact confirm the applicability and validity of the

model (de Mooij 2017). The model has also been criticised for the overly simplistic nature of its

dimensions (Osland and Bird 2000). Proponents of this model (de Mooij 2010) argue that it is the

simplicity of these dimensions that has resulted in the widespread adoption of this model. This

simplicity is argued to be straightforward and appealing to academic researchers. Ndubisi et al.

(2012) offer further support to this model by noting that these dimensions are conceptually

parsimonious and that they can be applied across a broad range of countries and cultures. These

authors argue that this model can be used as a starting point when attempting to understand

cultural differences; particularly in marketing relationships and relationship outcomes.

The efficacy of the scales used by Hofstede in developing the dimensions has been questioned by

some studies. Spector et al. (2001) and Spector and Cooper (2002) argue that the scales used in

32

the study have low internal reliability scores. They suggest that the construct validity of the scales

used to develop these dimensions is therefore questionable and that these scales need to be used

with caution in replication studies. This is a reason for concern as scales with low internal

reliability will not test a single homogeneous construct (Spector and Cooper 2002). This issue

appears to have been rectified as the more recent updates of the model; VSM 2008 and VSM 2013

report high internal reliability scores (Geert Hofstede 2018).

Comparisons have been made between Hofstede’s model and the GLOBE study (Smith 2006;

Venaik and Brewer 2008; Shi and Wang 2011). Some scholars; Yeganeh et al. (2009) even suggest

that the latter may replace the former in the future. However, the GLOBE Study has been criticised

by Hofstede et al. (2010); suggesting the possible threat this model poses to the hegemony of

Hofstede’s model (Siegfried 2014). The GLOBE model has been criticised less compared to that of

Hofstede possibly because the study is still receiving attention from researchers (Venaik and

Brewer 2016).

The most notable difference between the two models is the large number of organisations

surveyed in the GLOBE study compared to just one in Hofstede’s study. This has led to some

authors (Javidan et al. 2006) suggesting that the GLOBE study’s findings could be more

generalisable. The GLOBE study also provides two new dimensions; Performance Orientation and

Humane Orientation. The dimensions; Individualism versus Collectivism and Masculinity versus

Femininity have been broken down by the study to offer four new dimensions. As a result, the

GLOBE study can be argued to offer more avenues to explore cultural differences (Shi and Wang

2011).

The conceptual novelty of this study has however been questioned as it offers dimensions that are

similar to those offered by Hofstede’s model (Yeganeh et al. 2009); providing limited advancement

of knowledge compared to Hofstede’s dimensions (Magnusson et al. 2008). The GLOBE study

takes into account values as well practices as discussed earlier. This allows researchers to apply

eighteen dimensions to identify differences between supposed values and the extent to which

these values are reflected in practice. However, more recent research; Venaik and Brewer (2010)

shows that seven of the nine dimensions are in fact negatively correlated in terms of values and

practice; suggesting a case of diminishing marginal utility. Furthermore, the GLOBE study does not

offer scores for Sri Lanka whereas Hofstede’s dimensions can be used to compare the UK and Sri

Lanka.

33

Table 2.1 highlights the key differences between the constructions of the two studies

Differences GLOBE Model Hofstede Model Time Period 1994-1997 1968-1972 Primary Researchers Involved 170 1 Respondents Managers Managers & Non-Managers Organisations Surveyed 951 1 Type of Organisation Non-Multinational Multinational (IBM) Industries Food Processing, Financial and

Telecommunication Services Information Technology

Number of Societies/Countries

62 72

Analysis Team Effort Individual Effort Project Design American Dutch Number of Dimensions 9 4 (1980), 5 (1991) & 6 (2010) Internal Reliability High (House et al. 1999) Inconclusive (Spector et al. 2001;

Spector and Cooper 2002; Geert Hofstede 2018)

Table 2.1: GLOBE versus Hofstede (Adapted from Shi and Wang 2011 and modified by the author) High internal reliability scores have also been reported for the scales used in the GLOBE study

(House et al. 1999). Due to concerns about the reliability of the scales used by Hofstede (Spector

et al. 2001; Spector and Cooper 2002), researchers may be motivated to use the scales used in the

GLOBE study with greater confidence due to their allegedly greater construct validity. Having been

developed more recently, these dimensions may further ensure greater predictive validity

(Kirkman et al. 2006).

However, Hofstede’s model is used in this study and this is justified by the following points:

I. Hofstede’s model is the only model applied in multiple contexts covering consumer

behaviour, customer loyalty and loyalty programmes. Therefore, findings in this study can

be discussed in relation to previous research. The lack of studies using the GLOBE study in

these contexts therefore reduces its reference value

II. The GLOBE study does not offer a score for Sri Lanka which makes it difficult to make a

comparison with the UK. On the contrary, Hofstede’s model is in fact the only model that

provides country scores for Sri Lanka

34

Similarities and differences between the United Kingdom and Sri Lanka are shown in Figure 2.2

based on five of the six cultural dimensions; Hofstede et al. (2010). A score for Indulgence (IVR) is

not available for Sri Lanka to aid a comparison. Clear differences can be seen with regard to the

dimensions; Power Distance (PDI), Individualism (IDV) and Masculinity (MAS) while the two

countries have similar scores for the dimensions; Uncertainty Avoidance (UAI) and Long Term

Orientation (LTO).

Figure 2.2: Comparison of the United Kingdom and Sri Lanka (Hofstede Insights 2018) The following conclusions can be drawn based on these scores:

I. Power Distance: There is predicted to be greater and more equal distribution of power in

the United Kingdom with people expecting equal distribution of power. There will be

lower distribution of power in Sri Lanka with people focusing more on status

II. Individualism/Collectivism: Greater prominence will be given to individual needs and there

will be low levels of group integration and collectiveness in the United Kingdom. Collective

goals will be valued more in Sri Lanka with emphasis on group integration and cohesion.

There will also be closer ties amongst Sri Lankans while people in the United Kingdom will

value distance and privacy

35

89

66

35

51

6980

35

10

45 45

N/A0

102030405060708090

100

United Kingdom

Sri Lanka

35

III. Masculinity/Femininity: People in the United Kingdom will be more assertive and

competitive with less prominence given to building and maintaining relationships. Sri

Lankans will focus more on relationship building and value cooperation

IV. Uncertainty Avoidance: People in both countries will not be too risk averse and will not be

too sceptical of ambiguous situations

V. Long Term Orientation: People in both countries will focus on more immediate reward

oriented virtues with less focus on future oriented goals

2.4 Section Summary

This section has introduced the concept of national culture by synthesising the definitions offered

by various scholars. In doing so, it is highlighted that national culture not only binds people in one

society together; it also differentiates them from other societies. Furthermore, the influence of

national culture on consumer behaviour is highlighted. Through a review of three well-known

cultural models; Hofstede et al. (2010), Trompenaars (1993) and the GLOBE Study (House et al.

2004), common themes are identified amongst these models. The section is concluded with a

critique of these three models; leading to the selection and justification of Hofstede’s model to be

applied in this study.

The next section will review the industry structures within the UK and Sri Lanka using market data.

The data is then applied to structural theory to provide a more theoretically underpinned review

of the two markets.

36

Section 2

Industry Structure

37

2.5 Grocery Retailing in the United Kingdom

In 2017, the grocery retail sector accounted for nearly half of the entire UK retail market with a

value of £184.5 billion (IGD Retail Analysis 2017a). Despite its dominance, slower future growth is

expected with the sector estimated to be worth £212.9 by 2022; demonstrating an average annual

growth of approximately 2.88% (IGD Retail Analysis 2017a). Grocery retailing in the UK is

comprised of six main channels; hypermarkets, supermarkets, convenience, discounters, online

and other channels (IGD Retail Analysis 2017b). 77% of the sector is accounted for by

hypermarkets, supermarkets and convenience stores. Despite the dominance of these formats,

discounters and online formats are expected to experience significant growth in the next five

years; highlighting stagnation of the large retailing formats. The convenience sector is also

expected to enjoy future growth but the overall growth of modern grocery retailing appears to

come at the expense of traditional formats such as independent stores, markets and specialists

such as bakers and butchers. This sector is expected to grow by 3.6% by 2022 (IGD Retail Analysis

2017b).

Although significant growth in online grocery retailing is expected, this channel only accounts for

5.6% of the total grocery sector at present (IGD Retail Analysis 201b). The channel is predicted to

only make up 7.5% of the sector in 2022 and this is rather surprising given the advanced retail

structure in the UK. This sector should, in fact, be larger due to the high levels of internet

penetration in the UK; The World Factbook (2018a). Moreover, leading grocery retailers such as

Tesco, Sainsbury’s and Asda each have more than 88% online grocery retail coverage in the United

Kingdom (Mintel 2017a). The growth of the convenience and discount formats can be attributed

to the higher degree of importance placed by customers on convenience and lower base prices

when deciding where to shop (IGD Shopper Vista 2017a). The Everyday-Low-Price approach taken

by Aldi and Lidl has further heightened customer demand for lower prices. Overall, organised

grocery retailing constitutes 93.6% of the grocery sector at present; IGD Retail Analysis (2017b)

with the contribution expected to be 94.3% in 2022.

Channel 2017 £ billion 2022 £ billion % Change £ billion Hypermarkets 16.2 16.3 +0.6 Supermarkets 86.0 91.1 +5.9 Convenience 40.0 47.1 +17.8 Discounters 20.1 30.1 +49.8 Online 10.4 16.0 +53.8 Other 11.8 12.2 +3.4 Total 184.5 212.9 +15.4 Table 2.2: Channel Performance in UK Grocery Retailing (IGD Retail Analysis 2017b)

38

The UK grocery retail sector is comprised of at least twenty retail chains, operating over 57,500

stores; IGD Retail Analysis (2017b). Despite the multitude of grocery retailers, the sector is

dominated by the so called “Big Four” retailers; Tesco, Sainsbury’s, Asda and Morrisons. Tesco is

the clear market leader with a market share of 28% (Kantar Worldpanel 2018). The Big Four enjoy

over 70% share of the market. The oligopolistic nature of this sector was identified as early as the

1990s; Duke (1992) with subsequent studies; Burt and Sparks (2003) identifying the foothold

enjoyed by a few large scale retailers. Despite their dominance, the Big Four retailers have lost

market share and this is a result of the growth of Aldi and Lidl; contributing to the overall growth

of the discount channel. This is highlighted by the two discounters increasing their combined

market share by 13.5% between January and December 2017 whilst the Big Four have suffered

declines (Kantar Worldpanel 2018). The figure below confirms this; highlighting the increase in

market shares for Aldi and Lidl whilst the shares of the Big Four have declined.

Figure 2.3: Market Share Change (Adapted from Kantar Worldpanel 2018 and modified by the author) Another key aspect of the UK grocery retail sector is the dominance of own label products. Own

labels accounted for 57.1% of grocery retail sales in 2015 (Key Note 2015). High perceptions of

quality; 76% agreement reported towards these products highlights their success in this sector

(IGD Shopper Vista 2017b). The maturity of these products is evidenced through the tiered

0 5 10 15 20 25 30

Tesco

Sainsbury's

Asda

Morrisons

Aldi

Co-Op

Waitrose

Lidl

Other

28.2

16.7

15.5

10.9

6

6

5.3

4.4

6.9

28

16.4

15.3

10.7

6.8

5.8

5.2

5

6.8

Dec-17 Jan-17

39

strategies employed by the Big Four retailers in particular. Three tiers; premium, mid-range and

value ranges provide different price and quality points for customers (Key Note 2015).

2.6 Grocery Retailing in Sri Lanka

Whilst the Sri Lankan retail sector was valued at approximately £23.8 billion in 2016; LMD (2016),

the value of grocery retailing in the country is approximately £11.25 billion. Also known as general

trade, unorganised retailing accounts for approximately 84% of the entire sector with nearly

240,000 stores in operation (LMRB 2016). Unorganised retailing is comprised of independent

grocery retailers, independent grocery wholesalers and wet markets; local farmers’ markets and

fairs known to Sri Lankans as “Pola”. These retail formats are more prevalent in rural areas of the

country where approximately 95% of grocery retailing is carried out through such formats; driven

by the fact that 82% of Sri Lankans live in rural areas (The World Bank 2016). Organised retailing;

also known as modern trade is more prevalent in urban and suburban areas of the country but

only constitutes approximately 16% of the sector’s value (LMRB 2016).

The organised retail sector is valued at approximately £1.8 billion and despite being dwarfed by

traditional retailing formats; this format is growing rapidly and is expected to grow by 5% year-on-

year in the coming years (LMD 2016). This sector is comprised of five retail chains; operating

approximately 850 stores in total. The clear market leader is Cargills which enjoys more than 40%

market share whilst the other four retailers share 56% of the sector. In total, these five retailers

control 98% of the market with the balance 2% being shared by small or standalone supermarkets;

operating approximately 720 stores around the country. The table below shows the market shares

of these retailers but also highlights the fact that whilst operating a similar number of stores,

Cargills and Sathosa (government run cooperative grocery retail chain) enjoy dissimilar shares of

the market.

Retailer Market Share (Value) Estimated Total Market Share (Value) Number of Stores

Cargills 42% 6.72% 3241 Keells 20% 3.2% 662 Arpico 16% 2.56% 413 Laugfs 12% 1.92% 254 Sathosa 8% 1.28% 4005 Other 2% 0.32% 856 Table 2.3: Market Shares of Sri Lankan Organised Grocery Retailers (LMRB 2016; LMD 2017) 1Cargills Ceylon PLC (2018); 2Keells Super (2018); 3Arpico (2018); 4Laugfs Holdings (2018); 5Financial Times (2018)

40

Since the end of the civil war in 2009, Sri Lanka has attracted attention from various global non-

food retailers such as Mango (Mango 2018), Tommy Hilfiger (Financial Times 2014) and Hugo Boss

(Daily News 2017). Therefore, it is plausible that grocery retailers would begin to focus on Sri

Lanka as a potential market. Online grocery retailing is near non-existent in the country and this is

highlighted by the market leader; Cargills not offering this service. Keells and Arpico offer online

groceries but their performance is unknown. With internet penetration at 32.1%; The World

Factbook (2018b), the current growth of this channel is not very promising. This could also be

influenced to some extent by the food culture in the country where cooking is mainly done from

scratch (Albala 2010); perhaps resulting in customers being averse to online groceries.

Compared to the UK, hypermarkets are not as prevalent in the country and the majority of trading

is carried out through medium sized supermarkets. The convenience sector is small but growing

with Cargills, Arpico and Laugfs operating just over 100 stores in total (Arpico 2018; Cargills Ceylon

2018; Laugfs Holdings 2018). Discount retailing as operated by Aldi and Lidl in the UK for instance

is not evident in Sri Lanka but the market leader; Cargills focuses on an EDLP approach (Cargills

Ceylon 2018). Keells Super (Keells Super 2018) and Arpico (Arpico 2018) on the other hand appear

to focus more on product quality and customer service at a slightly higher price. In contrast to the

UK, own labels do not enjoy a large share of the market and are relatively new to customers. The

low market share of these products appears to be due to low customer perceptions of quality and

satisfaction (LMRB 2016).

Loyalty programmes are used in a more advanced manner by UK grocery retailers; gathering

customer data and interacting with customers using such data. However, no loyalty scheme is

used in such a manner in Sri Lanka where such schemes are used to offer members discounts and

other benefits. Further, the extent to which customer data is gathered by such schemes is

unknown. The usage of customer data in the UK is highlighted by personalised communication

based on shopping behaviour and individual customer data. Furthermore, the relative immaturity

of such schemes in Sri Lanka is highlighted by the market leader; Cargills only launching their

loyalty scheme in 2014.

41

2.7 Comparison of the Grocery Retail Structures in the UK and Sri Lanka

Although the market data on the two countries provide a clear picture of their respective

structures, applying retail structure theory would enhance the understanding of these structures.

Alexander and Doherty (2009) discuss five attributes that can be used to gauge a country’s retail

structure. These are applied to the UK and Sri Lanka in relation to the available market data to

provide a more theoretically underpinned comparison of the two countries’ structures.

2.7.1 Enterprise Density

Retail density refers to the extent to which retail operations are carried out by large retailers

(Alexander and Doherty 2009). This can be measured by ascertaining the number of inhabitants

per retail outlet; a number greater than 200 denoting a developed structure and a number less

than 100 denoting the opposite (Alexander and Doherty 2009). This is calculated by dividing the

population of a country by the total number of retail outlets in that country.

Country Population Retail Outlets UK 64,769,452 (World Factbook 2018a) 87,141 (IGD Retail Analysis 2017b) Sri Lanka 22,409,381 (World Factbook 2018b) 241,500 (LMRB 2016)

As such, the retail densities in the UK and Sri Lanka are 761 and 92 respectively; indicating a highly

developed structure in the UK and a developing structure in Sri Lanka. This further highlights the

potential for organised retailing in Sri Lanka; particularly to international retailers who are willing

to enter and operate within such structural and distribution systems (Alexander and Doherty

2009).

2.7.2 Market Concentration

The market share enjoyed by one retailer or a group of retailers indicates the level of market

concentration within an industry (Alexander and Doherty 2009). Given Tesco’s market share of

28% and the combined share of over 70% enjoyed by the “Big Four” (Kantar Worldpanel 2018), the

UK grocery sector is highly concentrated. Despite the developing nature of organised retailing in

Sri Lanka, the market share of 42% enjoyed by the market leader; Cargills and the combined share

of 98% owned by the five retail chains (LMRB 2016) makes the Sri Lankan organised grocery sector

highly concentrated. However, in contrast to the UK, it can be argued that the high levels of

concentration in Sri Lanka are less worrying for international grocery retailers due to the potential

for organised retailing; with the sector only making up 16% of the market (LMRB 2016).

42

2.7.3 Product Sector Balance

For Alexander and Doherty (2009), this refers to the balance between different retail product

sectors. In the context of this study, the channel composition of the grocery retail sector can be

used to understand the level of development. The developed sector in the UK is highlighted by the

availability of five different organised retail channels in addition to the unorganised channel (IGD

Retail Analysis 2017b). On the contrary, the predominance of the supermarket channel in Sri Lanka

underscores the developing nature of the sector and the potential for development in terms of

other channels such as hypermarkets, convenience and discount channels.

2.7.4 Employment Profile

Developed retail markets would have official employment figures compared to developing ones

(Alexander and Doherty 2009); thus, making this aspect another indicator of structural

development. Given the predominantly organised nature of retailing in the UK (IGD Retail Analysis

2017b), official employment figures will be available as these are submitted by organised retailers.

On the contrary, such figures would be less reliable and not easily available in Sri Lanka. This is

identified by Alexander and Doherty (2009) as a characteristic of developing markets due to family

members being involved in unorganised retailing and the prevalence of peripatetic retailing. This

also helps explain the value of the overall grocery retail sector in Sri Lanka. Due to the use of

immediate family members and the difficulty of quantifying mobile trade, an official or more

accurate value may not be attainable.

2.7.5 Innovation Levels

The degree of innovation within a grocery sector also demonstrates its degree of development.

The UK grocery sector exhibits high levels of innovation due to reasons such as the advanced

application of loyalty programmes, online retailing, self-checkout and more novel technological

advancements such as self-scanning of products. On the other hand, innovation levels are low in

Sri Lanka due to reasons such as basic implementation and use of loyalty programmes, negligible

online retailing as well as a lack of novel technologies. However, this too offers international

retailers various opportunities as there is significant potential within the market for innovations.

43

2.8 Section Summary

This section has reviewed the market data on the industry structures within the UK and Sri Lanka;

providing an in-depth understanding of their current status in terms of development, competition,

key influencers and growth. The saturated nature of the UK grocery retail sector is clearly

highlighted with the growth of discount retailers such as Aldi and Lidl becoming apparent

compared to the more established retailers. The overall dominance of the organised sector in the

UK is also evident with several channels contributing to this. On the contrary, the developing

nature of this sector in Sri Lanka is clear. However, this highlights the potential for organised

retailing formats in the country. Finally, the application of Alexander and Doherty’s (2009) retail

structure theory further highlights the disparities between the two countries. Table 2.4

summarises the structural differences between the United Kingdom and Sri Lanka.

Description United Kingdom Sri Lanka Value of Grocery Retailing £184.5 Billion £11.25 Billion Per Capita Value of Grocery Retailing* GBP 2,848 GBP 500 Organised Retailing 93.6% 16% Unorganised Retailing 6.4% 84% Number of Outlets (Total) 87,141 241,500 Number of Outlets (Organised Retail) 57,555 1,500 Number of Outlets (Unorganised Retail) 29,586 240,000 Organised to Unorganised Store Ratio 1.95 : 1 1 : 160 Online Retailing (% of Value) 5.6% N/A Own Label Penetration High; Tiered Low; Basic Loyalty Scheme Implementation Advanced Basic *Value of grocery retailing divided by the population (The World Factbook 2018a; 2018b) Table 2.4: Structural Differences between the UK and Sri Lanka

2.9 Chapter Summary

The first section of this chapter provides an overview of various definitions of culture in the

context of human behaviour; highlighting its influence on consumer behaviour. A critical

evaluation of three models of national culture has identified a suitable model to be applied in this

study. The two countries are distinguishable based on Hofstede et al.’s (2010) national cultural

dimensions. Clear differences can be seen with regard to the dimensions; Power Distance (PDI),

Individualism (IDV) and Masculinity (MAS) while the two countries have similar scores for the

dimensions; Uncertainty Avoidance (UAI) and Long Term Orientation (LTO). The second section

provides an overview of the two countries’ grocery retail structures based on market data;

highlighting clear differences in development. The application of structural theory further enables

the understanding of these differences.

44

Chapter 3

Customer Loyalty in Grocery Retailing

45

3.1 Introduction

This chapter begins with an overview of the emergence of relationship marketing which identifies

the importance of customer loyalty. The theoretical concept of customer loyalty is then reviewed

followed by some of its key definitions. Models developed to segment customer loyalty are then

reviewed; leading to the selection of one model to be applied as the theoretical base for this

study. Customer loyalty in grocery retailing is then reviewed with emphasis on its key antecedents

as identified in the literature. Customer loyalty and loyalty programmes across national cultures

and industry structures is finally reviewed and the chapter concludes with a conceptual framework

which is a culmination of the key aspects discussed in Chapter 2 (National Culture and Industry

Structure) and in this chapter.

3.2 A Shift in Marketing Philosophy- Relationship Marketing

The concept of earning customers’ favour and loyalty by satisfying their needs and wants was not

unknown to merchants in the feudal past (Berry 1995). The emergence of the industrial era; mass

production and mass consumption gave way to the involvement of middlemen as a result of the

separation of producers and customers. Engaging in relationship oriented contact with their

customers became harder as a result of the expansion of the customer base (Gronroos 1995).

These factors led to companies becoming more transaction oriented; giving prominence to

promoting and selling their goods by acquiring new customers. The need to build lasting

relationships with existing customers became less important as a result of this transaction

orientated approach (Sheth and Parvatiyar 1995).

A shift in focus took place as companies began to understand the importance of building

relationships with customers (Sheth and Parvatiyar 1995). This was driven by intensifying

competition and the deregulation of industries (Berry 2002) which meant that companies were

compelled to focus on customer retention. As a result of this shift in focus, companies identified

that superior relationships with customers were crucial to create a competitive advantage (Palmer

1996). The economic benefits of retaining customers further heightened the need to engage in

relationship building efforts (Gronroos 1995). Technological advances enabled companies to hold

more information about their customers and use such information to get to know their customers

better (Palmer 1996). Figure 3.1 depicts this shift in relationship marketing.

46

Figure 3.1: The Re-Emergence of Relationship Marketing (Source- Author) Relationship marketing can be defined as the process of attracting customers as well as

maintaining and enhancing relationships with them (Berry 1983). The focal aspect of this approach

is customer retention which leads to customer loyalty (Hennig-Thurau et al. 2002). Whilst focusing

on relationship development, this approach seeks to ensure that objectives of the company as well

as the customer are met (Boedeker 1997). Relationship marketing requires the alignment and

synthesis of marketing, customer service and quality to strengthen customer relationships

(Christopher et al. 1991). Despite these definitions giving prominence to relationships with

customers, relationship marketing should not be confined to relationships with customers but

should also apply to other stakeholders such as employees, investors and suppliers (Christopher et

al. 1991; Morgan and Hunt 1994; Reichheld 1994). However, the key emphasis is on building,

maintaining and enhancing relationships with customers.

The process of relationship marketing can be explained using the concept of the “Loyalty Ladder”

(Christopher et al. 1991). Too many companies give greater emphasis to identifying prospective

customers and trying to convert such Prospects into Customers. As a result, little attention is paid

to generating repeat business. Relationship marketing is aimed at making new Customers regular

Clients and to gradually move these Clients up the loyalty ladder to become Supporters of the

company and its products and/or services. The final stage is where these Supporters become

Advocates and Partners, engaging in active and vocal advocacy. They perform an important role as

a source of reference; positive word of mouth for others further down in the ladder.

Relationship marketing has a positive influence on profitability which is required for long term

survival and growth (Reichheld et al. 2000; Nguyen and Mutum 2012). This is because loyal

customers tend to buy more, be less sensitive to price premiums and actively engage in positive

word of mouth generation (Reichheld 1993; Zeithaml et al. 1996). Word of mouth plays a crucial

Relationship Marketing

Transaction Marketing

Industrial era

Mass production & consumption

Middlemen

Growing customer base

Intense competition

Deregulation

Economic benefits

Technological advancements

47

role in relationship marketing as it is deemed to be more effective and credible compared to more

traditional information sources such as advertisements (Kitapci et al. 2013). Increasing profits

could be a result of increased revenues and reduced operating costs as well as reduced marketing

costs since retaining existing customers has been found to be considerably cheaper than acquiring

new ones (Dawkins and Reichheld 1990; Palumbo and Herbig 2000). Table 3.1 highlights the

various benefits of customer loyalty as noted in the literature.

Benefit Source(s) Loyal customers spend more since they tend to purchase more compared to less loyal customers

Knox (1998); Peppers et al. (2002); Harris and Goode (2004);

Loyal customers are cheaper to serve and are less price sensitive

Dawkins and Reichheld (1990); Reichheld and Sasser (1990); Rowley (2005)

Loyal customers are more profitable to serve Reichheld and Sasser (1990); Reichheld (1994) Loyal customers act as referrals and engage in positive word of mouth (act as credible sources of information for others)

Dick and Basu (1994); Shoemaker and Lewis (1999); Duffy (2003)

Loyal customers engage in repeat purchasing Dawkins and Reichheld (1990) Customer loyalty helps companies create and maintain a competitive advantage

Dick and Basu (1994)

Table 3.1: Benefits of Customer Loyalty (Compiled by the author)

3.3 Customer Loyalty as a Theoretical Construct

Scholarly interest in the area of customer loyalty appears to have been in existence as early as the

1920s (Copeland 1923). Early research into customer/brand loyalty viewed this concept from a

purely behavioural viewpoint. The purchase frequency of a brand (Brody and Cunningham 1968)

and possibility of purchase (Farley 1964) were two of the constructs used to measure customer

loyalty towards a brand. Various other behavioural measures have also been used to define

customer loyalty (Kumar and Shah 2004). Ehrenberg (1988) identified five such measures and they

are outlined below.

i. The percentage of customers buying a particular brand

ii. Number of units of the brand purchased by each customer

iii. The percentage of customers continuously purchasing the brand

iv. The percentage of customers who are solely loyal to the brand

v. The percentage of customers buying other (related) brands

48

Using purely behavioural aspects to measure loyalty may have been guided by early scholarly

thinking that what a customer thinks or what goes on in their mind should not be taken into

consideration (Tucker 1964). This may have been driven by the assumption that loyalty would only

be demonstrated through action (Enis and Paul 1970). These assumptions suggest that an

attitudinal component does not exist or does not matter in loyalty; behaviour alone reflects loyalty

towards a brand or store.

As research in this area continued to grow, this purely behavioural view of customer loyalty began

to come under criticism. Day (1969) was one of the first scholars to question the appropriateness

of equating behaviour patterns to loyalty; arguing that that behavioural measures cannot

distinguish between true loyalty and spurious loyalty. True loyalty is a combination of favourable

behaviour and attitudes. On the other hand, spurious loyalty is favourable behaviour such as

repeat purchasing without a favourable attitude. Purely behavioural measures do not explain the

reasons behind such behaviour (Dick and Basu 1994) and strong attitudinal loyalty is required for

true loyalty to be present (Uncles et al. 2003). This is because behaviourally loyal customers may

lack the attitudinal loyalty; implicit emotional factors such as satisfaction and commitment that

govern consumer behaviour required for true loyalty (Baloglu 2002). As opposed to true loyals,

spurious loyals lack an attachment to the brand. The importance of securing true loyalty is

supported by longitudinal studies (Bove and Johnson 2009) where truly loyal customers have been

reported to be the most stable over time compared to spuriously loyal customers. Table 3.2

highlights some of the key advantages of attitudinal loyalty.

Characteristic Source(s)

Customers with high attitudinal loyalty are less prone to be affected by negative comments about a brand

Ahluwalia et al. (1999)

A positive attitude ensures that buying behaviour will continue Amine (1998) Customers with greater attitudinal loyalty are almost three times more likely to remain loyal to a brand compared to less attitudinally loyal customers

Baldinger and Rubinson (1996)

Favourable attitudes lead to (repeat) purchasing behaviour Baldinger and Rubinson (1996); Broyles (2009)

Table 3.2: Advantages of Attitudinal Loyalty (Compiled by the author)

However, it is surprising that attitudinal aspects were excluded when defining and measuring

customer loyalty in these early seminal studies. This is because clear references are made to

attitudinal components that were deemed to have a bearing on how customers purchase branded

goods as early as the 1920s (Copeland 1923). Three attitudinal components; recognition (resulting

49

from previous favourable experience or convincing advertising), preference (recognition creates

preference for one brand in the presence of multiple brands) and insistence (the customer accepts

no substitute brand except in the case of an emergency) were identified and the following quote

describes how they influence purchase decisions;

“If the consumer has no familiarity whatsoever with the brand of product to be purchased,

the entire sales burden rests on the salesman in the store visited. If the consumer

recognizes the brand, the manufacturer of that brand has taken the initial step in

consummating the sale to the consumer. If the manufacturer has established consumer

preference, the sale has proceeded one step further. If the consumer has the attitude of

insistence, it remains merely for the retail salesman to close the sale” (Copeland 1923,

p.288)

However, purely attitudinal measures may also be misleading and could be inadequate to measure

overall loyalty. This is because a favourable set of attitudes towards a brand does not necessarily

mean that such attitudes will not be applied to other brands (Uncles et al. 2003). Similarly,

attitudinal loyalty alone is not adequate for a company if it does not drive behaviour (Kumar and

Shah 2004). This is because behaviour will bring returns to the company and attitudinal loyalty on

its own may bring limited or no tangible returns. Unlike behavioural measures, attitudinal

measures are self-reported expressions of preference (Dubois and Laurent 1999) and these could

be misleading if not accurately reported by customers. Therefore a purely attitudinal approach to

analyse loyalty cannot be considered to be effective; similar to a purely behavioural approach.

Based on this discussion, it is clear that customer loyalty should not be looked at from a single

viewpoint; behavioural or attitudinal. These measures should be combined to gauge customer

loyalty as a composite construct. Doing so will enable a greater understanding of customer loyalty.

Finally, this approach will allow companies to target the right customers and induce purchasing;

thereby create true loyalty (Bandyopadhyay and Martell 2007).

3.4 Defining Customer Loyalty

Suggesting that one measure alone would not be sufficient to measure a complex

multidimensional concept such as brand loyalty, Jacoby (1971) developed his definition by

combining both behavioural and attitudinal dimensions. Behavioural loyalty reflects customer

tendencies and is measured through two elements; previous purchases of a particular brand or

patronage of a particular store as well as the measurement of probabilities of purchases in the

50

future based on previous purchases. Attitudinal loyalty in contrast is a customer’s psychological

disposition towards a brand or store and involves the measurement of customer attitudes. These

measures encompass emotional and psychological attachments linked to loyalty (Jacoby and Kyner

1973). Based on these measures, Jacoby (1971) defined loyalty as follows;

“Brand loyalty is the biased (non-random), behavioral response (purchase) expressed over

time by some decision-making unit with respect to one or more alternative brands out of a

set of such brands, and is a function of psychological (decision making, evaluative)

processes” (p.2)

Oliver (1999) defined customer loyalty as a deeply held dedication to re-buy/reuse a favoured

product/service recurrently in the future. This would lead to recurring purchasing of the same

brand/portfolio despite circumstantial pressures and marketing endeavours to tempt switching

behaviour. Both measures; behavioural and attitudinal are included in this definition as in the

previous definition but it goes further by including a situational aspect. A loyal customer would

have a favourable attitude towards a product or service and will engage in repeat purchasing

behaviour without being affected by situational aspects.

Both of these definitions include behavioural and attitudinal measures and neither measure

appears to have been given greater weight. As a result, these definitions are better suited to

measure customer loyalty as opposed to definitions that use single dimensions. Uncles et al.

(2003) support Oliver’s (1999) definition as it incorporates a situational aspect. This is because

such an element would be useful when behavioural and attitudinal dimensions fail to measure

customer loyalty fully. The situational element however causes some concerns. This is because a

behaviourally and attitudinally loyal (truly loyal) customer may be compelled to purchase a

product or service if the preferred product or service is unavailable. It is likely that the customer

cannot wait for the preferred option to become available and has no option but to look for an

alternative. Such an event may deem a customer to be disloyal for selecting an alternative. This is

in contrast to where a customer switches because of a competitor’s efforts where such a customer

can be considered disloyal. A more comprehensive understanding can be gained if this customer’s

future purchasing behaviour and attitudes towards the usual product or service are analysed

against those towards the one off product or service. If the customer goes back to the usual

product or service, then this customer cannot be considered to be disloyal as opposed to the

customer who switched because of competitor efforts.

51

3.5 Segmenting Customer Loyalty

Even the early concepts of customer loyalty using purely behavioural measures appear to have

identified that not all customers are the same in terms of their loyalty behaviour. Such measures

identified monogamous (fully loyal) customers, polygamous (in-between) customers and

promiscuous (no loyalty) customers based on their behaviour (Uncles et al. 2003). Identifying

customer loyalty segments will allow companies to obtain a better understanding of the nature of

loyalty and design appropriate marketing initiatives (Rowley 2005). In doing so, companies will be

able to manage their customers more effectively (Knox 1998).

Dick and Basu (1994) identified four loyalty segments based on the level of relative attitude and

repeat patronage (Figure 3.2). Relative attitude is a customer’s attitude towards a brand or

company in relation to their attitude towards alternatives. Relative attitude is said to be more

indicative of repeat patronage than an attitude towards an entity taken in isolation (Dick and Basu

1994). The strength of an attitude and its differentiation can be used to gauge the level of relative

attitude. The four segments conceptualised are loyalty; high relative attitude and high repeat

patronage, latent loyalty; high relative attitude and low repeat patronage, spurious loyalty; low

relative attitude and high repeat patronage and no loyalty; low relative attitude and low repeat

patronage.

Loyalty Latent Loyalty

Spurious Loyalty No Loyalty

Figure 3.2: Relative Attitude-Repeat Patronage Loyalty Segments (Dick and Basu 1994)

Loyals are most desirable for a company as they have high relative attitudes due to greater

perceived differences between alternatives complemented by high repeat patronage. Latent loyals

are customers with high relative attitudes but low repeat patronage due to non-attitudinal aspects

such as subjective norms and situational effects being equally/more influential. Spurious loyals

have high repeat patronage but low relative attitudes because non-attitudinal aspects influence

behaviour more and/or due to lower perceived differences between alternatives. No loyalty is a

result of low perceived differences between alternatives and non-attitudinal aspects having a

greater influence on repeat purchasing. Based on the previous model of Dick and Basu (1994) and

Relative Attitude

Low

Repeat Patronage

High

High Low

52

focusing only on one category (loyalty- high relative attitude and high repeat patronage), Rowley

(2005) conceptualised subcategories to better understand these loyal customers (Figure 3.3). Her

argument was that these “loyal” customers deserved special attention as they are more important

for the future of a company. Using positive and inertial as differentiating factors, four

subcategories were conceptualised. They are Captives, Convenience Seekers, Contended and

Committed.

Captives are loyal customers who have an inertial attitude and continue to repurchase due to a

lack of choice. These customers are more common in business to business contexts due to high

switching costs and in some customer contexts such as banks where there are switching costs

involved. Convenience seekers are those who again have an inertial attitude but engage in

repurchasing mainly due to convenience. Contended customers have a positive attitude and may

even provide positive referrals but are inertial in their behaviour. These customers seem to be

more common in industries with additional services and add-ons. Committed customers are those

who are positive in both attitudes and behaviour. These customers engage in positive word of

mouth generation and also engage in repeat purchasing.

Captive Contended

Convenience Seeker Committed

Figure 3.3: Subdividing “Loyals” (Rowley 2005) The models of Dick and Basu (1994) as well as Rowley (2005) use attitudinal and behavioural

components to categorise customer loyalty. Such loyalty segments can be considered to be more

accurate as opposed to those developed using behavioural or attitudinal measures in isolation. In

comparison, the first loyalty segmentation matrix (Dick and Basu 1994) can be used as a first step

when identifying types of loyalty to obtain a general overview of the customer base. The second

segmentation matrix (Rowley 2005) can be applied to complement the first as it only looks at one

particular loyalty segment in greater detail.

Ignoring less loyal segments and giving more attention to the most loyal segment could be unwise.

This is because a company should not just be content with its “loyal” customer segment but it

Behaviour

Positive

Inertial

Attitude Inertial Positive

53

should also strive to make customers in other segments become more loyal. It is important for

companies to continuously compare the size of each loyalty segment over time to understand how

customers shift between segments (Baloglu 2002). In doing so, companies can try and attempt to

identify reasons for such shifts so that they can take appropriate action. A segmentation matrix

similar to that of Dick and Basu (1994) will allow a company to capture its entire customer base

and analyse the extent to which loyalty segments are positively or negatively affected by their

general marketing and relationship marketing initiatives.

A similar loyalty segmentation matrix named The Diamond of Loyalty was developed by Knox

(1998) with three underlying principles;

I. Most customers buy on a portfolio basis

II. All customers are not equal

III. Loyalty is retention with attitude

By combining the purchase portfolio (number of brands purchased) and degree of involvement

(with a company or brand) of a customer, Knox (1998) divided customers into four broad loyalty

categories (Figure 3.4). Customers who purchase from a narrow portfolio (more repeat purchasing

of one or few brands with a higher share allocated) and have a high level of involvement

(attitudinal attachment) with a brand were called Loyals. The second loyalty category was called

Habituals. Although these customers also purchase from a narrow portfolio, they have a low level

of involvement and are likely to switch if their usual brand(s) is unavailable as their behaviour is

governed by pure habits, not attitudes. Variety Seekers and Switchers both purchase from a wide

portfolio (less repeat purchasing of one or few brands with a lower share allocated) with the

differentiating factor again being the level of involvement. Variety seeking customers have a

higher level of involvement with the variety of brands they purchase while switchers have a low

level of involvement and will act opportunistically as they are interested in price deals and

discounts.

54

Figure 3.4: Diamond of Loyalty (Adapted from Knox 1998 and modified by the author) Knox’s (1998) model shares certain similarities with Dick and Basu’s (1994) model. Customers are

divided into broad categories based on the level of purchasing and involvement. This method of

categorisation allows a company to take its entire customer base into account. However, this

model looks at purchasing from a portfolio perspective and not from a single brand or store’s

perspective as in Dick and Basu’s model. Furthermore, Dick and Basu’s model looks at relative

attitudes while this model looks at attitudes in isolation. This could be a shortcoming as relative

attitude could be more indicative of customer loyalty behaviour (Dick and Basu 1994).

Another model, developed by Oliver (1999) looks at loyalty developing in a sequential pattern over

four phases with the first three phases being attitudinal and the fourth being behavioural in nature

(Figure 3.5). The essence of this model is that while customers can be loyal at each of the three

attitudinal phases, the degree of loyalty increases as a customer advances through each phase.

The four phases are Cognitive Loyalty, Affective Loyalty, Conative Loyalty and Action Loyalty.

Cognitive loyalty is the first phase where a particular brand may be preferred by a customer based

on information about the brand or prior experience and knowledge about the brand. Loyalty at

this phase is said to be shallow with a low level of satisfaction. In the second phase; affective

loyalty, a liking or favourable attitude towards the brand is developed through higher levels of

satisfaction resulting from cumulative satisfactory usage of the brand. Although the degree of

loyalty at this stage is greater than that in the previous phase, customers are likely to switch as

satisfaction alone does not guarantee loyalty. Conative loyalty is the third phase where the

customer develops a behavioural intention to repurchase the brand resulting from repetitive

positive attitudes towards the brand. The customer holds a deep commitment (motivation) to

Variety Seekers

Switchers

Habituals

Loyals

Low

High Low

High

Customer Involvement Purchase Portfolio

55

repurchase the brand. Action loyalty is the final phase where a behavioural element comes into

play where the previous commitment to repurchase is converted to actual repurchase behaviour.

Figure 3.5: Oliver’s Four Stage Loyalty Model (Oliver 1999) This model takes into account both attitudinal and behavioural aspects to define the stages of

loyalty, similar to the models discussed previously. However, the previous models categorise

customers using matrices based on their attitudinal and behavioural characteristics. This model

uses a sequential process through which a customer is said to progress with the degree of loyalty

increasing at each stage; resulting in repurchase behaviour. Further, attitudinal and behavioural

aspects are not combined in any of the stages. As a result, instead of placing customers in different

categories, this model identifies the stage where a customer is located. The first stage is

concerned with information regarding a brand that creates customer preference towards the

brand. The second stage takes satisfaction as the primary factor resulting in loyalty. This is because

the main focus of Oliver (1997) was on satisfaction’s impact on loyalty. This will not be an ideal

indicator as Oliver (1999) later argues that satisfaction alone does not create loyalty. The level of

satisfaction will govern the purchase intentions in the third stage resulting in actual repeat

purchasing in the final stage.

Compared to the previous matrix based models, this model could provide different results and a

different way to analyse and segment customer loyalty. Using a matrix based model together with

this sequential model may enable a company to obtain both a deeper and wider understanding as

opposed to using either a matrix based model or this sequential model in isolation. Although the

researcher would have preferred to use these two different types of models to obtain a more

comprehensive picture of customer loyalty in the two countries, it should be noted that comparing

findings may have led to confusions as the measurements are conceptualised differently as

evident through the review of these models.

Despite being applied in previous grocery retailing research (Omar and Sawmong 2007), Oliver’s

(1999) framework may lack relevance in the context of this study due to its sequential nature;

Beliefs, Affect, Intentions and Action (Blut et al. 2007). This is because this model does not

combine behavioural and attitudinal elements when measuring the four stages. Therefore, this

Phase 1: Cognitive

Loyalty

Phase 2: Affective Loyalty

Phase 3: Conative Loyalty

Phase 4: Action Loyalty

56

study uses Dick and Basu’s (1994) loyalty framework as it incorporates the concept of relative

attitude. This element also makes the framework stand out from the other matrix based models;

Rowley (2005) and Knox (1998) as it enables a researcher to understand different levels of

attitudinal loyalty across brands or retailers.

3.6 Customer Loyalty in Grocery Retailing

The importance of securing customer loyalty to grocery retail stores was identified over fifty years

ago (Tate 1961) primarily due to increasing direct competition between grocery retailers as a

result of structural developments. Tate reported that even in 1961, only 10% of families studied in

the USA were loyal to one grocery retailer and this too was purely due to a lack of choice. The

study also revealed that more than 33% of these families patronised five or more grocery retail

stores; suggesting a low level of true loyalty and a high level of promiscuity. The need to improve

store loyalty was highlighted by findings that store loyal customers are more profitable to a

grocery retailer since these customers allocate the majority of their shopping budget to the focal

retail store (Cunningham 1961). Consequent studies (Enis and Paul 1970; East et al. 1995; Knox

and Denison 2000) as well as more recent ones (Sreedhara and Babu 2010) have corroborated this

finding. The various advantages of customer loyalty as identified previously have also driven

grocery retailers to focus on engendering greater levels of customer loyalty.

3.6.1 Measuring and Defining Customer Loyalty in Grocery Retailing

Measuring customer loyalty to grocery stores became imperative for grocery retailers. The earliest

studies used purely behavioural dimensions to measure grocery retail loyalty. Tate (1961) used the

number of stores visited and the proportion of the food budget spent at the favourite store to

measure the level of grocery loyalty. His findings suggested that when customers patronised more

stores, the proportion of their spending in their main store reduced from 100% in a perfectly loyal

scenario to 53% when seven stores were patronised by a customer. Cunningham (1961) argued

that the number of stores was not relevant and instead, research should be focused on the

proportion of spend allocated to each store thus, identifying what he termed “first store loyalty”.

Enis and Paul (1970), alluding to these studies and several others, used three measures to study

grocery store loyalty in the USA;

i. The proportion of the budget allocated to the focal store during a specific period ii. Number of stores patronised during this period iii. Number of store switches or changes

57

Their research findings suggested that most loyal customers spent on average 92 cents out of

every dollar spent in their first choice store. This reduced to 73 cents, 57 cents and 46 cents

respectively as the level of loyalty decreased. Dunn and Wrigley (1984) used Cunningham’s (1961)

measure of first store loyalty as the basis for their study on store loyalty for grocery products and

provided similar findings. All these approaches have used purely behavioural measures to study

grocery store loyalty. The influence of purely behavioural measures has survived with more recent

studies using such measures (Knox and Denison 2000).

As discussed previously, purely behavioural measures are inadequate as they exclude attitudinal

elements that allow researchers to identify underlying reasons behind overt loyalty behaviour;

“Behavioural conceptualisations and operationalisations are often inadequate to explain

how and why store patronage occurs” (Bloemer and de Ruyter 1998, p.500)

Further, even though repeat visiting behaviour is important, retailers should pay more attention to

the antecedents that influence the occurrence of such behaviour (East et al. 1995). Based on this

argument and the one above and adapting Jacoby’s (1971) definition of brand loyalty, Bloemer

and de Ruyter (1998) offer a definition of store loyalty that includes both behavioural and

attitudinal measures;

"The biased (i.e. non-random) behavioural response (i.e. revisit), expressed over time, by

some decision-making unit with respect to one store out of a set of stores, which is a

function of psychological (decision making and evaluative) processes resulting in brand

commitment" (Bloemer and de Ruyter 1998, p.500)

Similarly, Noordhoff et al. (2004) conceptualised store loyalty as a combination of three elements.

Whilst incorporating proportion of visits paid to a store (share of visits) and the proportion of a

customer’s budget spent in a single store (share of wallet), they also added the customer’s bias

towards the focal store as a function of a mental process consisting attitudinal preferences and

commitments towards the store. In a more recent study, Omar et al. (2010, p.13) defined store

loyalty as follows where both behavioural and attitudinal elements were included. This definition

also includes a situational element as included in Oliver’s (1999) definition of brand loyalty;

“High positive attitudes and repeat purchase behavior towards a particular store thereby

causing repetitive purchasing in the same retail store, despite situational influences and

marketing efforts by competitors in persuading switching behavior”

58

It is evident that using both behavioural and attitudinal elements to measure and define store

loyalty provides a more comprehensive understanding as opposed to using such elements in

isolation. Based on these definitions and also incorporating the relative attitude element of Dick

and Basu (1994), the researcher defines grocery retail store loyalty as follows.

“A mixture of behavioural (relatively higher share of visits and/or wallet and positive word

of mouth) and attitudinal loyalty (greater positive relative attitude) towards a grocery

retailer in spite of situational factors and competitor efforts to induce switching behaviour”

3.6.2 Segmenting Customer Loyalty in Grocery Retailing

The importance of segmenting customers based on their loyalty towards grocery retailers has

been identified by several authors. Store loyalty as a segmentation criterion could be more

effective than traditional methods of market segmentation as loyalty based market segmentation

is associated with a retailer's profit making capabilities (Enis and Paul 1970). This is justified by

findings discussed previously that loyal customers are more profitable. Furthermore, segmenting

customers based on their store patronage characteristics may enable retailers to adopt a focused

approach of targeting such customers that will result in enhancing customer loyalty and store

performance. This is more effective compared to non-selective strategies that fail to identify

customers with greater potential for loyalty (Baltas et al. 2010).

The models used to segment customer loyalty as reviewed previously can be used to segment

customer loyalty towards grocery retailers. As discussed previously, models that use both

behavioural and attitudinal measures are likely to be more effective. Oliver’s (1999) framework

has been applied in the context of grocery retailing in the UK (Sawmong and Omar 2004; Omar

and Sawmong 2007). The models of Dick and Basu (1994), Knox (1998) and Rowley (2005) have

not previously been applied in grocery retailing contexts. However, Dick and Basu’s (1994) model

has received empirical support in a customer product setting where the presence of the four

loyalty segments was confirmed (Jensen 2011); confirming the validity of this model. Furthermore,

the applicability of this model in grocery retailing has been supported by more recent studies

(Ngobo 2017).

Rowley’s (2005) conceptual model may not be as applicable to a grocery retail setting as it only

looks at loyal customers and ignores other loyalty categories. This model can be used to

complement a more general segmentation model. The Diamond of Loyalty (Knox 1998) can also be

59

used in segmentation efforts. Purchase portfolio (number of brands purchased) should be

replaced by “store portfolio” (number of stores patronised) so that identified segments reflect

store loyalty, not loyalty to brands in a portfolio. Using the “store portfolio” measure would be

more appropriate as a majority of grocery shoppers have been found to shop on a portfolio basis

(Seiders and Tigert 1997; East et al. 1998; Luceri and Latusi 2012).

Another crucial difference is evident between Oliver’s (1999) four-stage model of customer loyalty

and the matrix based models. Scores of each stage can be linked to individual customers

(Sawmong and Omar 2004; Omar and Sawmong 2007). Matrix based models on the contrary will

look at customer loyalty by combining behavioural and attitudinal measures and categorise

customers accordingly. A customer will not be in more than one category as a result. This is a

critical factor as this allows the identification of different customer loyalty segments. Dick and

Basu’s (1994) loyalty typologies will be used in this study and the use of one such framework is

important to ensure consistency and clarity. This model is more suitable as it incorporates relative

attitudes; allowing the identification of loyalty segments based on behaviour as well as attitudes in

relation to other grocery retailers. The four typologies will be used for the discussion with one

modification; True Loyalty will replace Loyalty (Figure 3.6). This is to ensure clarity of discussion

where True Loyalty refers to high repeat patronage and relative attitude as defined by previous

research (Day 1969; Baloglu 2002; Uncles et al. 2003).

Figure 3.6: Relative Attitude-Repeat Patronage Loyalty Segments (Adapted from Dick and Basu 1994 and modified by the author) 3.6.3 Antecedents of Customer Loyalty in Grocery Retailing

Factors that impact grocery retail loyalty need to be identified as this will allow retailers to focus

their attention on such factors to improve levels of customer loyalty. Enis and Paul (1970) when

segmenting customer loyalty in the United States identified several determinants of store loyalty.

Their results suggest that store loyalty could be influenced by customer socioeconomic

True Loyalty Latent Loyalty

Spurious Loyalty No Loyalty

High Low

Relative Attitude (Attitudinal)

Low

Repeat Patronage (Behavioural)

High

60

characteristics such as education level and occupation. High store loyalty was observed to a

greater extent in less educated respondents. The same was observed in terms of occupation

where households headed by labourers or farm workers were more store loyal as opposed to

those whose heads worked in professional, technical or managerial capacities. Their findings did

not provide evidence to suggest that psychological characteristics impacted customers’ store

loyalty. Since then, many studies have been carried out to identify determinants of store loyalty.

The key antecedents identified in the literature are discussed below.

3.6.3.1 Store Image

Store image was first conceptualised by Martineau (1958) as a store’s personality that would be

formed in the mind of a customer which in turn would influence the customer’s shopping

behaviour. The following definition of store image reflects this view; “the complex of a consumer’s

perceptions of a store on different (salient) attributes” (Bloemer and de Ruyter 1998, p.501).

Martineau (1958) identified store layout and architecture, symbols and colours, advertising and

sales personnel as determinants of a store’s personality. The nature of a store’s image in the mind

of a customer is determined by how these salient attributes are evaluated and weighted against

each other (Doyle and Fenwick 1974). Since its conceptual inception in 1958, researchers have

identified an almost exhaustive list of attributes that constitute store image. These attributes or

characteristics also known as the retail mix of a particular store contribute towards the overall

image of that store (Bloemer and de Ruyter 1998). Table 3.3 highlights retail mix elements

identified by various authors over the years since its conceptual inception in 1958.

Author(s) Retail Mix Elements Martineau (1958) Store layout and architecture, symbols and colours, advertising and sales

personnel (4 elements) Lessig (1973) Store impression, variety, pleasantness of customers, produce, honesty,

prices, meats and service (8 elements) Lindquist (1974) Merchandise, service, clientele, physical facilities, comfort, promotion, store

atmosphere, institutional and post-transaction satisfaction (8 elements) Doyle and Fenwick (1974)

Product, price, assortment, styling and location (5 elements)

Bearden (1977) Price, quality of merchandise, assortment, atmosphere, location, parking facilities and friendly personnel (7 elements)

Ghosh (1990) Location, Merchandise, store atmosphere, customer service, price, advertising, personal selling and sales incentive programs (8 elements)

Chang and Tu (2005)

Facilities, store service, store activities and convenience (4 elements)

Thomas (2013) Price, product assortment, product quality and store service (4 elements) Table 3.3: Attributes of Retail Mix or Store Image (Compiled by the author)

61

The relative importance of these elements has been ranked differently in various studies and one

universally accepted ranking sequence is not available. As a result, it would be interesting to

explore whether customers from culturally and structurally disparate countries would view the

importance of these elements differently. Identifying any possible differences would enable

grocery retailers to focus on the most important elements in each market to improve their overall

store image. This is because improving store image has a positive impact on store loyalty (Chang

and Tu 2005; Nesset et al. 2011; Imran et al. 2013; Das 2014).

3.6.3.2 Relationship Quality

Relationship quality is the overall assessment of a relationship's strength; the extent to which it

meets the expectations and needs of the parties involved based on a history of successful or

unsuccessful encounters or events (Crosby et al. 1990). Relationship quality is determined by three

elements; the degree of trust placed on one another by parties involved in a relationship, how

satisfied these parties are with the relationship and the extent to which these parties are

committed to the long term maintenance of the relationship (Smith 1998). These three elements

are the most widely cited in retail studies (Athanasopoulou 2009) and are thus reviewed in this

section.

Trust Considered to be an essential requirement for a successful relationship (Morgan and Hunt 1994;

Berry 1995), trust is defined as one party’s confidence to rely on the exchange party (Moorman et

al. 1993). Trust is present when one party is confident about the exchange party’s integrity and

reliability (Morgan and Hunt 1994). These two statements highlight the importance of confidence

and reliability as components of trust (Garbarino and Johnson 1999). Credibility and benevolence

have also been linked to trust in previous studies (Vesel and Zabkar 2010). Credibility is when the

exchange party carries out its role effectively and reliably whilst honouring promises. Benevolence

refers to the exchange party being concerned for the other party’s wellbeing and best interests

while refraining from attempting to exploit the relationship.

High levels of trust lead to more durable and high quality relationships (Nguyen and Mutum 2012)

and the creation of a competitive advantage (Morgan and Hunt 1994). Trust has also been found

to positively influence relationship commitment (De Wulf and Odekerken-Schroder 2003; Wang

2008) as well as play a mediating role between satisfaction and commitment in a relationship

(Guenzi et al. 2009). More importantly, trust has been found to positively influence store loyalty

(Shpëtim 2012).

62

Satisfaction Satisfaction can be defined as a customer’s evaluation of the perceived difference between a

product’s expected level of performance and its actual level of performance (Tse and Wilton

1988). This evaluation leads to an outcome of pleasure or displeasure (Oliver 1999). In a more

service centric context, satisfaction is a customer’s post consumption evaluation of a service

(Ganguli and Kumar 2008). In a general sense, satisfaction can be looked at as a post consumption

experience which compares expected quality and perceived quality (Sivadas and Baker-Prewitt

2000).

The extent of a customer’s satisfaction or dissatisfaction with a product or service will therefore be

determined by the actual perceived standard or quality of a product or service meeting, exceeding

or falling below the expected standard or quality of that product or service. Satisfaction is an

important determinant of relationship quality as it has been found to have a positive impact on

trust (Yap et al. 2012) and customer loyalty (Oliver 1999). Satisfaction has also been found to

generate a positive relative attitude towards a product or service (Sivadas and Baker-Prewitt

2000).

Commitment The common theme emerging from the definitions of commitment is that a committed party in a

relationship will make an effort to maintain and prolong the relationship due to the belief that the

relationship is important (Wong and Sohal 2002). Commitment can be in three forms; emotional

commitment, calculative commitment and normative commitment (Vesel and Zabkar 2010).

Emotional commitment refers to a customer’s affect-laden, uninfluenced desire to maintain a

relationship with a company. On the contrary, calculative commitment is more logical and rational

in nature and is determined by the importance placed on the relationship by a customer in

monetary terms. Normative commitment is free of any intentions of personal gains where a

customer is guided by what is appropriate and honourable. Commitment is an important

dimension as it helps gauge the strength of a relationship and measure customer loyalty (Wong

and Sohal 2002) by allowing a company to identify “stayers” and “leavers” (Wilson 1995). More

importantly, research findings have confirmed that commitment has a positive impact on

customer loyalty (De Wulf and Odekerken-Schroder 2003; Wang 2008; Ou et al. 2014).

63

3.6.3.3 Service Quality

Service quality, in brief, is the perceived difference between a customer’s expectation of a service

and the actual level of service received (Gronroos 1984; Parasuraman et al. 1985). Various

measurement tools such as the service quality gap model (Parasuraman et al. 1985), service

quality dimensions (Parasuraman et al. 1988) and service performance model (Cronin and Taylor

1992) were developed to help service oriented companies measure their levels of service quality.

These tools were developed because service quality was identified to have a significant impact on

customer satisfaction (Oliver 1999) and customer loyalty (Zeithaml et al. 1996). Studies conducted

by Martinelli and Balboni (2012) and Shpetim (2012) produced results to suggest that service

quality positively impacted store loyalty through store satisfaction as well as store trust in the

latter study. A direct positive impact of service quality on store loyalty was found in studies

conducted by Odekerken-Schroder et al. (2001), Molina et al. (2009) and Shpëtim (2012).

Figure 3.7 summarises the key antecedents of customer store loyalty. Store Image, through its

various attributes has been found to influence customer store loyalty (Chang and Tu 2005; Nesset

et al. 2011; Imran et al. 2013; Das 2014). Similarly, relationship quality has been found to impact

customer loyalty and this is influenced by trust (Shpëtim 2012), satisfaction (Oliver 1999) and

commitment (De Wulf and Odekerken-Schroder 2003; Wang 2008; Ou et al. 2014). The final

antecedent of service quality is another crucial element influencing customer loyalty (Odekerken-

Schroder et al. 2001; Molina et al. 2009; Shpëtim 2012).

Figure 3.7: Key Antecedents of Customer Store Loyalty

Customer Loyalty

Store Image

Relationship Quality

Service Quality

64

3.7 Loyalty Programmes in Grocery Retailing

A loyalty programme is a collection of marketing initiatives aimed at improving customer loyalty

(Sharp and Sharp 1997; Leenheer et al. 2007). These tools are used by retailers to offer benefits to

their customers to create, enhance and maintain customer loyalty (Palmer et al. 2000; Yi and Jeon

2003; Garcia-Gomez et al. 2006). These programmes are built on the premise of building personal

relationships with customers where two-way-communication is a key focus (Meyer-Waarden and

Benavent 2009). All these definitions except for one; Sharp and Sharp (1997) focus on customer

loyalty from a purely behavioural perspective in respect of such programmes. Such measures were

discussed to be inadequate when attempting to understand customer loyalty. Based on these

definitions and considering the composite nature of customer loyalty; behavioural and attitudinal,

the author defines a loyalty programme as;

“An integrated system which uses individual customer information to develop personalised

marketing strategies allowing the retailer to engage in two way communication and offer

customised rewards to customers in an attempt to create attitudinal (greater relative

attitude) and behavioural (repeat purchasing, greater proportion of spending and positive

word of mouth) loyalty towards the retailer”

Loyalty programmes are viewed by several scholars (De Wulf et al. 2001; Filipe et al. 2017) as a

relationship marketing tool, crucial to retailers’ pursuit of customer loyalty. Such initiatives are

grounded in the theories of reciprocity (Huppertz et al. 1978) and perceived relationship

investment (Smith 1998). The theory of reciprocity argues that customers would act in a

favourable manner or develop positive emotions towards a retailer if that retailer has expended

time, effort or any other irrecoverable resources towards customers (Bagozzi 1995). Such actions

from a retailer are argued to result in customers “perceiving” the retailer to have invested in the

relationship, resulting in favourable attitudes (Hart and Johnson 1999) and reciprocity (Smith and

Barclay 1997). More recent research has identified that perceived relationship investment in the

form of loyalty programmes would positively influence relationship quality (Mimouni-Chaabane

and Volle 2010) as well as customer loyalty (Huang 2015); outcomes which are in fact, reciprocal

action from customers.

The increasingly ubiquitous nature of these loyalty programmes is evidenced by their usage in

various retail contexts (Mimouni-Chaabane and Volle 2010), including grocery retailing (Filipe et al.

65

2017). Whilst there is some debate on the effectiveness of such initiatives in engendering

customer loyalty, there is widespread evidence to support the positive relationship between

loyalty programmes and customer loyalty. For instance, loyalty programmes have been found to

positively influence both behavioural and attitudinal loyalty (Garcia-Gomez et al. 2006; Omar et al.

2013). Further, their influence on both short-term as well as long-term customer loyalty has been

confirmed (Lewis 2004; Liu 2009).

From a development perspective, Green Shield Stamps can be considered the earliest form of

loyalty schemes in grocery retailing with the first stamps in the United Kingdom emerging in the

1960s (Retrowow 2017). These stamps were given to customers by grocery retailers in return for

their purchases and these stamps were pasted in books which customers were able to exchange

for gifts. The late 1970s saw the decline of green shield stamps. This decline was due to their

negative impact on profits and shoppers focusing solely on tangible incentives attached to such

stamps as well as many grocery retailers offering these stamps; resulting in the impact of this

initiative on any individual retailer declining (Boukhobza 2005).

The 1990s saw the emergence of electronic points based loyalty schemes that are still in existence

with many of Europe’s largest supermarkets adopting them. This was driven by their need to

enhance their knowledge of their customers (Mauri 2003). Loyalty schemes were also

implemented to increase repeat purchasing and obtain a greater proportion of spending from

customers (Meyer-Waarden and Benavent 2009). Most retailers have continued their loyalty

schemes with others adopting such schemes due to the common belief that such schemes can be

used to retain customers (Demoulin and Zidda 2009).

Leading grocery retailers in the UK such as Tesco and Sainsbury’s have adopted such schemes.

These schemes are used to discourage customer switching whilst gathering customer purchase

data in order to offer focused promotions and rewards. Other grocery retailers such as Morrisons

and Waitrose have also launched loyalty programmes. Tesco and Sainsbury’s use a range of

partner merchants in their loyalty programmes; providing customers a wide range of redemption

options. Overall, the loyalty programme industry in the UK is mature with several grocery retailers

offering such programmes. However, it should be noted that Asda does not offer a loyalty

programme having discontinued its scheme in 2000 (BBC 2010) and instead focuses on an “Every

Day Low Price” approach. Discount retailers; Aldi and Lidl also appear to favour the low price

approach over loyalty schemes.

66

On the contrary, loyalty programmes in Sri Lanka are still growing with only two retailers; Keells

and Arpico operating established loyalty programmes. Compared to loyalty programmes run by

Tesco and Sainsbury’s, these schemes are used merely to offer members discounts although some

additional redemption options are offered through partner merchants. Another key difference is

the lack of individualised customer contact and customised offers although customer shopping

behaviour is tracked. The relative immaturity of such schemes is highlighted by the market leader;

Cargills only launching its loyalty scheme in 2014. This scheme also offers discounts to customers

in return for their membership.

3.7.1 Loyalty Programme Rewards

All loyalty programmes offer some sort of reward to members but the intentions of offering such

rewards could be different. Some of these intentions are to make customers use the card so that

their shopping behaviour can be tracked, to try and make cardholders behave in a way the retailer

expects or simply to reward them for repeat patronage (Mauri 2003). Rewarding customers is

crucial as benefits have been found to have a positive impact on customer satisfaction (Berenguer-

Contri et al. 2009) which is linked to store loyalty. It is important for retailers to understand

reward types preferred by customers so that such rewards can be used to create greater customer

satisfaction, resulting in higher levels of store loyalty. Different types of rewards or benefits have

been identified by previous scholars and these are reviewed below.

Hard versus Soft Benefits: Hard benefits are those that are economic in nature such as discounts,

coupons, points and rebates (Bridson et al. 2008). Such benefits are easier for customers to

evaluate but can be easily imitated by competitors (Gable et al. 2008). Soft benefits on the other

hand do not generally carry any monetary value but are highly valued by customers (Bridson et al.

2008). Such benefits could take the form of a bouquet on a special occasion to after-hours

shopping or special invitations. Such rewards are harder to implement but also harder for

competitors to imitate. These benefits are more likely to create an emotional bond between the

customer and the loyalty programme (Gable et al. 2008). This is because they are more

experiential in nature; making them more attractive to customers (Cedrola and Memmo 2010). It

is important for retailers to identify the type of benefits individual customers prefer so that such

benefits could be used more effectively (Omar et al. 2010).

67

Timing: The timing of rewards refers to whether a customer receives a reward immediately

following a purchase; immediate reward or after some time; delayed reward (Keh and Lee 2006; Yi

et al. 2013). Immediate rewards for example are offered for every store visit while delayed

rewards are offered for every nth visit (Yi and Jeon 2003). Whilst immediate rewards can be used to

induce customer switching from competitors, delayed rewards could be used to reward

customers’ future purchases (Zhang et al. 2000).

Direct versus Indirect: Direct rewards are directly linked to the value proposition of a product;

related to the product purchased while indirect rewards are not linked to the product purchased

(Dowling and Uncles 1997).

Aggregated versus Segregated: Aggregated rewards carry greater value as they are integrated

and are offered together. Segregated rewards are less valuable since they are offered in separate

parts. There is no agreement on the use of these rewards as customers have been found to have

different preferences towards these rewards (Yi et al. 2013).

Tiered Schemes: Tiered schemes offer customers differential benefits and rewards based on their

status as defined by various predetermined tiers (Dreze and Nunes 2009). Whilst such

programmes are more pervasive in high involvement contexts such as airlines, their suitability in

low involvement contexts such as grocery retailing has been questioned (Arbore and Estes 2013).

3.7.2 Concerns Related to Loyalty Programmes

Even though loyalty programmes are being used by many grocery retailers, such programmes have

raised several concerning questions. These are discussed below.

Customer Loyalty to the Programme or the Store: One of the objectives of using loyalty

programmes as discussed above was the need to retain customers and increase customer loyalty.

Several studies have argued that loyalty programmes may increase loyalty towards the

programme rather than the retailer (Yi and Jeon 2003; Leenheer et al. 2007; Evanschitzky et al.

2012) and this is more prevalent in low involvement contexts such as grocery retailing (Dowling

and Uncles 1997). This raises the question whether customers who are loyal to the loyalty

programme would defect if the loyalty programme is terminated. Therefore, it is important that

retailers should attempt to direct the attention of customers to their core products/services rather

than the rewards/offers (Yi and Jeon 2003).

68

Opportunistic Behaviour: Retailers are said to receive a higher share of wallet from programme

members than from non-members (Bridson et al. 2008). However, this could be misleading

because such members may opt to join such loyalty schemes simply to obtain the benefits offered

to them (Leenheer et al. 2007). It is also likely that customers may switch as they receive better

offers from other programmes (Meyer-Waarden and Benavent 2009). Asda, Aldi and Lidl serve as

good examples where these grocery retailers focus on an “Every Day Low Price” approach instead

of implementing loyalty programmes.

Opportunity Costs: Loyalty programmes are expensive to setup and maintain (Dowling and Uncle

1997). These costs range from sunk costs incurred when establishing the infrastructure for such

programmes to rewards and benefits offered to customers. Incurring such high costs creates an

opportunity cost for retailers. This is because the investment could be directed towards other

marketing initiatives and retailers may not always be able to determine which investment would

offer better returns.

3.8 Customer Loyalty across Cultures

Cross-cultural research has been carried out to assess the impact of culture and cultural

differences on consumer behaviour. Such research has been driven by the pivotal role culture has

been found to play in shaping the way consumers make purchase decisions (Bahhouth et al. 2012;

Lindridge et al. 2014), process information (Arnold and Bianchi 2001) and behave (Ackerman and

Tellis 2001; Luna and Gupta 2001; Mazaheri et al. 2014).

Despite the vast number of studies that have been carried out in the area of cross-cultural

consumer behaviour, only a few studies have focused on the impact of such differences on

customer loyalty in general (de Mooij 2004; Lam 2007; Abubakar 2013). More focused studies in

the sphere of grocery retail store loyalty remain scarce (Straughan and Albers-Miller 2001; Khare

2012). A review of the literature published between 1970 and 1990; Sojka and Tansuhaj (1995)

identified only one study where culture is linked to brand loyalty. This study; Saegert et al. (1985)

compared two ethnic groups in the United States of America. This intra-country study does not

report any significant differences between the two ethnic groups; Hispanic and non-Hispanic in

terms of unwillingness to try new products and previously purchased product recall. Since then, a

small number of studies focusing on the impact of such differences on customer loyalty in general,

using Hofstede’s dimensions have emerged. Table 3.4 below summarises some of these studies

and their similarities/differences are discussed thereafter.

69

Author(s) Context Findings Lam (2007) Studying the impact of national

culture on customers’ brand loyalty proneness in Australia

Customers with high Individualism and high Uncertainty Avoidance have a greater proneness to be brand loyal

Seock and Lin (2011)

Study of young customers’ loyalty towards apparel retail stores in the USA and Taiwan

Customers from countries with high Collectivism (low Individualism) have greater loyalty tendencies

Ndubisi et al. (2012)

Studying the impact of Uncertainty Avoidance on customer loyalty in the banking sector in Turkey and Malaysia

Turkish customers (high Uncertainty Avoidance) were found to be more loyal than Malaysian customers (low Uncertainty Avoidance). Uncertainty Avoidance has a positive impact on customer loyalty

Table 3.4: Studies on Cross-Cultural Customer Loyalty using Hofstede’s Dimensions

The first and third studies provide similar findings in relation to Uncertainty Avoidance. These two

studies report that customers from high Uncertainty Avoidance cultures are more loyal. These

findings can be supported by Hofstede et al.’s (2010) argument that customers in high Uncertainty

Avoidance countries tend to avoid ambiguities and limit risks. However, the third study; Ndubisi et

al. (2012) is perhaps more valid as it compares two nations’ cultures as recommended by Hofstede

et al. (2010). The first study; Lam (2007) on the other hand compares customers within one

country and this approach is questionable since Hofstede’s dimensions of national culture should

be applied at country level and not at the individual level (Hofstede 2002). However, different

scores for a particular dimension are possible within a country as not all people would rate these

dimensions the same (Triandis 1989).

Lam (2007) also reports that Individualism has a positive impact on brand loyalty proneness.

However, the reported statistical significance is greater than 0.05; thus, the likelihood of this

finding being a product of chance cannot be ruled out. The second study by Seock and Lin (2011)

reports that customers with high Collectivism have greater loyalty tendencies. Their hypothesis is

based on the notions presented by Hofstede (1980; 1991); conforming to group norms,

suppressing internal beliefs and behaving in socially acceptable ways. These findings provide

conflicting and perhaps confusing insights to researchers due to their varied nature. Therefore,

studying the direct impact of national culture may not be appropriate. Instead, incorporating other

elements or even looking at how national culture mediates relationships would perhaps provide a

clearer and greater understanding.

70

3.9 Cross-Cultural Retail Store Loyalty and Loyalty Programmes

In a study of the USA, Australia, France and South Korea; Straughan and Albers-Miller (2001)

identified differences in loyalty towards domestic and international retailers. Their findings

highlight that high Collectivism has a positive influence on domestic retailer loyalty. This finding

suggests that Sri Lankan consumers would be more loyal to domestic retailers due to their greater

Collectivism. This may have a negative effect on foreign retailers attempting to expand into such

countries. The hypothesis used in their study is grounded in the conceptualisation of this

dimension (Hofstede 1980; 1991). Studies conducted in India (Khare 2012; Khare et al. 2014;

Pandey et al. 2015) report that consumers are more loyal to local stores. This phenomenon is

linked to the high Collectivism and low Masculinity in India where customers tend to focus on

group cohesion and relationship building (Hofstede et al. 2010). Due to cultural similarities with

India, this finding could be used to speculate that Sri Lankan customers may also be more loyal to

their local stores compared to UK customers. Earlier research; Goldman (1974) also suggests that

customers in countries such as India and Sri Lanka are more likely to maintain personal

relationships with their local traditional retailers. While these findings suggest that customers in

countries such as Sri Lanka would be more loyal to domestic retailers and local stores, they do not

identify the effect of national culture on loyalty to grocery retailers within a country.

The impact of national culture on loyalty programmes remains sparsely explored (Mattison

Thompson and Chmura 2015). Previous studies have not focused on how they are received by

customers in countries with different national cultures and what these customers expect from

such programmes. In a study of restaurant loyalty programmes in the USA; Park et al. (2013)

identified that less Long Term Oriented customers were positively affected by monetary (hard

benefits) and immediate rewards. Their hypothesis is based on the premise that such customers

may prefer short term gains due to their short term oriented cultural values as mentioned by

Hofstede and Bond (1988); Hofstede et al. (2010). This finding is applicable to the United Kingdom

and Sri Lanka as both countries have low Long Term Orientation scores (Hofstede Insights 2018). It

is likely that customers in these two countries may also prefer such rewards from grocery retail

loyalty programmes.

71

A study of FMCG purchasing behaviour in India, China, Russia, Brazil and Germany; Mattison

Thompson and Chmura (2015) yielded the following findings;

i. Customers from high Power Distance and low Individualistic countries prefer loyalty

programmes that offer related (direct) rewards

ii. Customers in low Masculine and low Uncertainty Avoiding countries prefer unrelated

(indirect) rewards

iii. Customers from high Masculine and high Uncertainty Avoiding countries reject loyalty

programmes and prefer immediate promotional offers instead

These findings provide some conflicting insights as to how loyalty programmes should be used in

countries with different national cultures. According to the first finding, Sri Lankan customers; high

Power Distance and low Individualism may prefer related rewards. On the other hand, the second

finding suggests that customers with low Masculinity and low Uncertainty Avoidance may prefer

unrelated rewards. These two findings are contradictory as Sri Lankan customers can be

speculated to prefer both related and unrelated rewards due to their similar scores. The third

finding can be applied to the UK; customers characterised by high Masculinity may reject loyalty

programmes and prefer immediate promotional offers instead. While these two studies provide

insightful findings; the application of the first is questionable since it is applied within a single

country and does not compare more than one country as recommended by Hofstede (2002).

However, these findings could be used as points of reference in this study. Overall, these findings

clearly show that a universal loyalty programme strategy may not be effective in countries with

different national cultures.

The review of customer loyalty and loyalty programmes across national cultures clearly highlights

the lack of focused studies in the area. This is typified by the following statement which specifically

focuses on customer loyalty;

“Culture dimensions still remains, as the greatest problem for international retailers in

diversifying into foreign countries. Thus, no previous research has been able to establish a

perfect model relating to culture that could be used in analyzing cultural norms” (Sawmong

2006, p.2)

72

3.10 Structural Impact on Customer Loyalty in Grocery Retailing

The review of the structure in the two countries highlights clear disparities and the two countries

occupy the two ends of the retail development scale. While the UK is one of the countries in the

first wave of modern food retailing with a highly developed structure, Sri Lanka represents the

third wave with predominantly traditional food retailing formats (Reardon et al. 2012). Despite the

clear structural differences between such countries, only a few studies have focused on the impact

of such differences on grocery retail customer loyalty. Luceri and Latusi (2012), in a study of

grocery retail patronage behaviour in Italy, identified that a customer’s patronage set; the number

of stores used widens as the number of alternative stores increases. This finding has some

implications in the context of this study. It suggests that there could be more store switching

behaviour in the UK due to the higher number of grocery retail chains in operation. Such

behaviour in grocery retailing is more likely as switching barriers are relatively low compared to

other retail industries such as banking (El-Manstrly et al. 2011; Tesfom and Birch 2011; Valenzuela

2014). Such behaviour in Sri Lanka could also become more prevalent as more retailers commence

operations or enter the country (Seiders and Tigert 1997). Structural developments may also result

in such behaviour. This is supported by previous research; Tate (1961) where increasing

competition and structural developments were identified to induce greater customer promiscuity.

Seminal work in this area; Goldman (1974) notes that customers in countries with under-

developed retail formats such as Sri Lanka are more likely to maintain personal relationships with

their local traditional retailers. Such behaviour is reported by more recent studies in India; Khare

(2012), Khare et al. (2014) and Pandey et al. (2015). Goldman (1982) also reports “selective

adoption” of modern retailing formats in developing countries. This phenomenon lends support to

the notion that customers in countries such as Sri Lanka may rely on traditional stores for most of

their grocery needs. However, such behaviour is very likely to have changed since these early

studies with the retail structure in the country developing (LMRB 2016). Nevertheless, these

studies are inconclusive and do not shed adequate light on the possible impact of industry

structure on customer loyalty to grocery retailers.

73

3.11 Literature Review Summary & Conceptual Framework Development

This review highlights the lack of studies in the area of grocery retail customer loyalty and loyalty

programmes. The structural impact of grocery retailing on customer loyalty also remains thinly

explored. Through a review of several models of national culture, Hofstede et al.’s (2010) model

was selected to form the basis of discussion for the rest of this research. The choice of this model

is justified by its widespread use and the availability of cultural scores for both countries. Clear

differences as well as similarities in national culture exist between the two countries based on

these dimensions. Significant differences between the two countries in respect of industry

structure are also evident as evidenced by the review of market data from the two countries and

the application of industry structure theory.

The importance of relationship marketing to retailers is discussed; highlighting the various benefits

of customer loyalty. The value of using a composite measure of customer loyalty; behavioural and

attitudinal is discussed following the review of the literature. Such an approach allows a more

holistic view to be obtained. The choice of a customer loyalty model depends on its ability to

segment customers by differentiating amongst various customer loyalty segments. Dick and Basu’s

(1994) model was selected to act as the base for this research due to two reasons. Firstly, it allows

the categorisation of customers based on their loyalty. Secondly, the model incorporates the

concept of relative attitude; a concept not used by other segmentation models.

Relationship marketing has been adopted by grocery retailers in their attempts to secure greater

customer loyalty and one of the most prominent tools used by such retailers is loyalty

programmes. However, research has not been carried out to examine how culturally different

customers respond to such loyalty initiatives. The influence of industry structure also remains

unexplored. This lack of knowledge thus raises the question of if and how loyalty programmes

should be implemented in countries with different national cultures and industry structures.

Despite the lack of previous research, a conceptual framework is presented in Figure 3.8. Although

simplistic, this framework was used when developing the discussion guide for the first primary

research phase; consumer focus group discussions. The components of this framework were used

as key areas to investigate during these discussions. The four typologies presented by Dick and

Basu (1994) are at the centre of this framework; representing the concept of customer loyalty.

This model was selected as the base for this research as justified in sections 3.5 and 3.6.2. The key

antecedents of customer loyalty; Store Image, Relationship Quality and Service Quality as

reviewed in section 3.6.3 are included together with loyalty programmes as reviewed in section

74

3.7. These four elements are depicted within the four boxes with arrows directed towards

customer loyalty. These arrows highlight the influence of these elements on customer loyalty as

identified in the aforementioned sections.

National Culture and Industry Structure are depicted as surrounding these elements. Due to the

lack of existing knowledge, it is speculated that these two elements will have a direct influence on

customer loyalty as well as an indirect impact through the four influencing elements. Hofstede et

al.’s (2010) model is applied in this study and its selection was justified in section 2.3.5 in Chapter

2. Its widespread use in consumer behaviour research makes it a suitable reference point when

discussing the findings from this research. Furthermore, cultural dimension scores for both the UK

and Sri Lanka are only offered by this model; thus allowing a proper comparison of the two

countries. The five cultural dimensions; Individualism (IDV), Power Distance (PDI), Masculinity

(MAS), Uncertainty Avoidance (UAI) and Long Term Orientation (LTO) as identified in Figure 2.2 in

Chapter 2 comprise national culture.

Figure 3.8: Conceptual Framework (section numbers are highlighted within brackets) Dick and Basu’s (1994) customer loyalty typologies are depicted at the centre of the diagram

Relationship Quality

(3.6.3)

Service Quality (3.6.3)

Loyalty Programmes (3.7)

Store Image (3.6.3)

True Loyalty

Spurious Loyalty

Latent Loyalty

No Loyalty

National Culture: PDI---IDV---MAS---UAI---LTO

Industry Structure

75

Chapter 4

Methodology

76

4.1 Introduction

The literature review highlighted the lack of previous research focusing on the impact of national

culture and industry structure on grocery retail customer loyalty. The lack of research in relation to

loyalty programmes is also evident. Primary research was required to generate new knowledge

given the lack of extant literature. The methodological approach adopted for primary research is

discussed in this chapter. The chapter commences with a discussion of the research philosophy

used to guide the methodology. The research approach adopted is then discussed, followed by an

illustration summarising the key steps in the research process. Each stage within this process is

discussed in detail with justifications provided for the choice of data collection methods used,

sampling procedures and analysis techniques followed.

4.2 Research Philosophy

A research philosophy or paradigm is a way of looking at the world and is defined by its

epistemological and ontological assumptions, the role of values (axiology) and the corresponding

methodology (Doyle et al. 2009; Teddlie and Tashakkori 2010). Despite the availability of several

research paradigms to modern day social scientists, Positivism and Interpretivism are considered

the two main paradigmatic stances (Johnson and Onwuegbuzie 2004). Positivism assumes that

knowledge is derived from observable phenomena, where the focus is on causalities and law-like

generalisations. This view results in the truth, with the nature of reality being objective as it is

external and independent of social actors and prescribes that values of the researcher do not

interfere in the research (Bryman and Bell 2011; Saunders et al. 2012). Interpretivism by contrast

contends that there are multiple realities (Doyle et al. 2009). These are subjective and socially

constructed because knowledge is derived from subjective meanings and details of specifics

behind phenomena (Saunders et al. 2012). The researcher’s values play a role because of their

immersion in the research (Doyle et al. 2009).

Given the antithetical nature of these two philosophies in terms of their epistemological and

ontological assumptions as well as their axiological considerations (Doyle et al. 2009; Bryman and

Bell 2011; Saunders et al. 2012), it could be argued that they can only be applied to purely

quantitative or qualitative research studies. In research studies where the research question(s)

cannot be sufficiently answered through a purely quantitative or qualitative approach, such purist

philosophical stances may not be applicable (Teddlie and Tashakkori 2009). Given the qualitative

nature of objective two; Refine the conceptual framework by exploring the impact of national

77

culture and industry structure on grocery retail customer loyalty in the UK and Sri Lanka and the

quantitative nature of objective three; Develop a confirmatory framework by further investigating

the impact of national culture and industry structure on grocery retail customer loyalty in the two

countries, a purist philosophical stance could not be adopted in this study. Therefore, this research

was carried out from the philosophical standpoint of Pragmatism which endorses the combined

use of quantitative and qualitative data through mixed methods research (Teddlie and Tashakkori

2009). This enables the benefits of both methods to be utilised in a way that complements

strengths and avoids overlapping weaknesses of the individual methods used (Johnson and Turner

2003).

Pragmatism can be defined as a "deconstructive paradigm” that rejects concepts such as "truth"

and "reality"; instead, this paradigm focuses on "what works" with regard to the research

questions being investigated (Tashakkori and Teddlie 2003).

“Pragmatism rejects the either/or choices associated with the paradigm wars, advocates

for the use of mixed methods in research, and acknowledges that the values of the

researcher play a large role in interpretation of results" (Tashakkori and Teddlie 2003,

p.713).

Unlike the philosophical positions espoused by Positivism and Interpretivism which appear to be

deeply grounded in the abovementioned epistemological, ontological and axiological assumptions;

Pragmatism posits the notion that priority should be given to the research question(s) (Masters et

al. 2006; Lawal 2009; Biesta 2010).

4.3 Mixed Methods Research

Mixed methods research is where a researcher collects and analyses data, integrates the findings

and makes inferences based on both qualitative and quantitative methods within a single study

(Tashakkori and Creswell 2007). Mixed methods research is regarded as the third major research

approach after quantitative and qualitative research (Johnson et al. 2007) and has emerged as a

research movement with a recognised name and distinct identity (Denscombe 2008). Despite the

rising popularity of this type of research, quantitative and qualitative purists have advised against

the mixing of quantitative and qualitative paradigms and their associated methods. The

Incompatibility Thesis; Howe (1988) is one such argument which suggests that mixing qualitative

78

and quantitative methods should be avoided and such combinations are epistemologically

incoherent. Guba (1990, p.81) followed with the following contention:

“Accommodation between paradigms is impossible... we are led to vastly diverse,

disparate, and totally antithetical ends”

Despite such opposition, there is evidence to suggest that mixed methods research has been used

for a long time in the background without being acknowledged or considered as part of a new

paradigm (Denscombe 2008). Furthermore, this research approach has been used in ethnographic

studies for more than a century (Guest 2013). As mixed methods research continued to progress

greatly over the years (Bryman 2007), previously opposing scholars have gone on to accept the

possibility of mixing the two paradigms previously deemed incompatible (Guba and Lincoln 2005).

This is because mixed methods research is regarded as the solution to the paradigm wars.

Paradigm wars are described as the long enduring, circular and profoundly unproductive debates

surrounding the advantages and disadvantages of quantitative versus qualitative research (Feilzer

2010). Mixed methods research allows researchers to mix and match design elements that offer

the best chance of answering their specific research questions (Johnson and Onwuegbuzie 2004).

In so doing, this approach allows researchers to use methods that complement individual

strengths and avoid overlapping weaknesses (Johnson and Turner 2003).

An exploratory sequential mixed methods design was used in this study where a qualitative phase

precedes a quantitative phase. The qualitative phase was used inductively to explore and generate

new data which could then be deductively tested to ascertain their generalisability. The primary

purpose of an exploratory design is to use the findings from the first qualitative phase to guide the

design of the ensuing quantitative instrument. This is to ensure that the constructs tested are

representative of customer perceptions, grounded in their own vocabulary. This also allows the

previously obtained findings from a small number of participants to be generalised using a larger

sample (Creswell and Plano Clark 2011). Similarly, such a design can be used to develop a

quantitative instrument with good psychometric properties; acceptable reliability and validity by

analysing the previously gathered qualitative data (Creswell 2013). Therefore, such a design’s use

could be twofold. Firstly, to develop new, emergent data when existing knowledge is scarce and

secondly, to develop a quantitative instrument based on the qualitative data (Creswell and Plano

Clark 2011). In the case of this research, both outcomes are relevant.

79

Figure 4.1: Main Stages in the Research Process

Conclusions and Recommendations (Objective 5)

Discussion of Findings & Development of Theoretical Framework (Objective 4)

Quantitative Data Analysis & Interpretation of Findings Confirmatory Framework Development

Primary Research Stage 2: Empirical Study 2- Questionnaires (Objective 3)

Development of the Survey Instrument

Thematic Analysis & Interpretation of FindingsIdentification of Questionnaire Content Refine Conceptual Framework

Primary Research Stage 1: Empirical Study 1- Three Focus Group Discussions each in the UK and Sri Lanka (Objective 2)

Conceptual Framework Development & Identification of Gaps in Knowledge

Literature Review (Objective 1)National Culture and Industry Structure Customer Loyalty

The Impact of National Culture and Industry Structure on Grocery Retail Customer Loyalty

Lite

ratu

re R

evie

w u

pdat

ed a

nd re

fined

ove

r the

cou

rse

of th

e Ph

D

80

Figure 4.1 illustrates the research process followed in this study. A critical review of the existing

literature on national culture and customer loyalty was carried out as secondary research along

with a review of market data on the two countries’ industry structures. This review resulted in the

identification of a suitable cultural model and structural elements that differentiate the two

countries. The review of extant literature on customer loyalty, more specifically in the areas of

grocery retailing and loyalty programmes highlighted the lack of studies focusing on the impact of

cultural and structural differences resulting in the need for primary research. Primary research was

carried out in two main stages encompassing two empirical studies.

The first stage was carried out inductively through qualitative research methods. Given the limited

nature of prior knowledge, qualitative research through its exploratory approaches enables

researchers to identify new themes and generate new knowledge through data (Patton 2002). A

qualitative approach in this stage was more suitable compared to a quantitative approach as the

former could provide an in-depth and detailed understanding of the phenomena under study and

enable the researcher to arrive at concepts and theories inductively through the collected data

(Bryman and Bell 2011). This approach further enables the understanding of the meanings

attached to such phenomena by participants, grounded in their own vocabulary and how they

perceive such phenomena (Creswell 2014). The development of the quantitative instrument was

informed by the data collected at the previous inductive stage; the most pertinent qualitative

findings were used in the development of the survey.

The mixed methods research design used in the primary research stage is illustrated in Figure 4.2

as per the accepted notation standard (Creswell and Plano Clark 2011). The two strands were

given equal status; QUAL and QUAN (Johnson et al. 2007) as the qualitative phase was used not

only to develop the quantitative instrument but also to develop new data. Similarly, the

quantitative phase was used not just to generalise the findings from the previous qualitative phase

but also to statistically test the refined conceptual framework developed through the analysis of

the qualitative data to empirically confirm previously conceptualised theory.

Figure 4.2: Research Design

Empirical Study 1

QUAL 6 Focus Groups (32 participants)

3 Each in the UK & SL

Empirical Study 2

QUAN 550 Questionnaires

274 in the UK & 276 in SL

81

4.4 Empirical Study 1: Focus Group Discussions

Six focus groups (three in each country) were conducted in the United Kingdom and Sri Lanka.

Theoretical saturation was achieved after preliminary analysis of three discussions in each country.

Focus group discussions are regarded as the most effective qualitative data collection tool to

generate ideas and probe subject areas from a number of different angles and identify

perspectives that could not have been previously identified by the researcher (Krueger and Casey

2009). Furthermore, in relation to participant observations and individual interviews, focus groups

were selected due to the following reasons:

I. Focus groups are not stronger than participant observations and individual interviews in

observing interactions and probing of informant knowledge respectively but they are

better at combining these two components than the other two techniques would in

isolation; offering the best of both worlds (Morgan and Spanish 1984)

II. The interaction among participants makes it easier for them to consider and reflect on

aspects of their daily life that are usually taken for granted (Acocella 2012); such as in

this study, grocery shopping and loyalty programme usage

III. Given the rather mundane nature of grocery shopping behaviour, interactions within

groups were important to ensure that participants kept sharing their ideas, thoughts and

experiences (Morgan 1997). This would not have been possible with individual interviews

as individual interviewees may not have been able to generate data to such an extent

without such interactions

IV. The group process of a focus group also enables participants to explore and clarify their

opinions in ways that are less easily accessible in individual interviews (Kitzinger 1995),

leading to a multiplication of information (Acocella 2012)

Despite their selection, certain limitations of focus groups were identified and steps were taken to

mitigate these limitations. Firstly, a lack of control is attributed to focus groups due to group

interactions (Morgan 1998). However, the moderator ensured that the discussions remained

focused on the research questions by using a more directive stance in such situations (Fern 2001).

Dominant and withdrawn participants are also problematic as the former would tend to control

the discussion while the latter may not contribute adequately (Fern 2001). Having developed a

rapport with each participant at the beginning of the focus groups (Greenbaum 2000), the

moderator was able to engage withdrawn participants and politely keep dominant participants

under control. As with other types of data collection, social desirability bias was identified as

82

influencing the validity of the findings generated through the focus group discussions (King and

Bruner 2000). However, this was not considered a critical issue due to the non-sensitive nature of

the research.

Focus group discussions were convened in a relaxed setting that allowed participants to sit around

a large table. The researcher moderated all six discussions with a trained observer, impartial to the

research acting as an assistant. The researcher remained passive during the discussions, using a

directive approach only to keep the discussions on track. Whilst clearly communicating with

participants verbally, the researcher also used non-verbal communication to facilitate the

discussion. Body language and gestures were used to express interest in a topic, to get participants

to keep talking or even as a sign to dominant participants. Hand gestures were used to encourage

participants to keep talking, get other participants to contribute or even to signal that the

moderator would like to move the discussion forward to a different area. Furthermore, facial

expressions such as a simple smile or nodding were used to encourage participants and to show

support. Prompts such as ‘yeah’, ‘uh-huh’, ‘hmm’… were also used. The observer’s role was rather

limited; helping the researcher follow the discussion guide and pointing out any participants

whose attempts to contribute to a given conversation may have been missed.

4.4.1 Focus Group Content and Discussion Guide

A structured discussion guide was developed before conducting the focus group discussions to

ensure that each focus group followed a similar format and that the relevant research objective

was met. The conceptual framework (Figure 3.8) was used as the basis for the discussion guide.

The discussion guide was developed as per the guidelines of Krueger (1998) and Krueger and

Casey (2009). A pilot study was conducted iteratively to ensure the questioning route was logical

and understood well using a convenience sample in each country. Further, this ensured that the

right questions were asked, they were asked in the right sequence to ensure optimum collection

of insights and questions were grounded in the participants’ own vocabulary; ensuring clear and

consistent understanding. Although Trust and Commitment are captured in the conceptual

framework, these concepts were not included as they were not appreciated by participants during

the pilot studies. Participants failed to elaborate on these two concepts; Commitment in particular

while Trust was considered by them as a ‘given’; resulting in a lack of elaboration. A copy of the

discussion guide can be found in Appendix A.

83

The discussion guide was not followed strictly, rather allowing the discussion to run uninterrupted

to ensure the unhindered flow of insights from participants. The moderator prompted and probed

only when required to encourage participants to engage in the discussion and to articulate their

thoughts. Relaxed conversational and non-technical language was used to ensure a free-flowing

discussion and a comfortable environment for the participants. Care was taken to ensure that all

participants were provided equal opportunities to speak and none of the participants dominated

the discussions. Participants who seemed to be quiet, shy and withdrawn were encouraged to

engage in the discussion.

Participants were welcomed with biscuits and fruit juice in the UK while Sri Lankan participants

were treated to local savouries and tea/coffee. Once everyone had settled down and taken their

seats, the nature of the focus group was explained to them briefly. The moderator stressed that

there were no right or wrong answers. Participants were then asked to introduce themselves

which served two purposes. Firstly, it allowed the participants to get to know one another.

Secondly, this allowed the researcher to identify each participant when transcribing the

discussions. Before asking the first question, each participant was asked to write down and share

with the rest of the group what they had most recently purchased. This enabled the moderator to

focus the attention of the first question towards grocery shopping and ensured that each

participant could easily contribute early in each group; creating an inclusive and relaxed

atmosphere. This writing activity helped participants feel relaxed and engage in casual

conversation which served as an icebreaker.

The discussions covered three broad sections. The first section was used to identify grocery

shopping habits of the participants and to subtly probe the reasons for patronising certain grocery

retailers over others. The second section was used to explore elements valued by participants

when shopping for groceries and to get participants to engage in a friendly debate over these

factors. The third and final section asked participants to share their thoughts and opinions about

their patronage of retailers and their perceptions of loyalty programmes. Each group discussion

lasted approximately 80 minutes; average length 82 minutes in the UK and 79 minutes in Sri Lanka.

4.4.2 Sampling

Purposive sampling is recommended when collecting qualitative data as the purpose is of primary

importance (Morgan and Scannell 1998; Krueger and Casey 2009). This is contrary to quantitative

data collection methods where random sampling is given prominence to ensure that each

84

potential respondent within a research population has an equal and known chance of being

selected due to the importance of generalisability (Krueger and Casey 2015). Therefore, potential

participants were identified with the main purpose of the focus group discussions in mind; to

explore their perceptions towards customer loyalty and loyalty programmes. Participants were

recruited from a pool of personal contacts known to the researcher and preliminary screening was

done to ensure that they were regular shoppers and responsible for the significant part of grocery

shopping in their household.

Sixteen participants comprising a broad age range known to the researcher were used for the

three focus groups in the UK and they were unfamiliar with the research. All participants studied,

worked or lived in Bournemouth. Given the ideal number of five to eight participants per group for

non-commercial purposes (Krueger and Casey 2009) and similar recommendations by other

scholars (O’Brien 1993; Morgan 1997; Crowley and Gilreath 2002), a minimum of five participants

were secured for each discussion. Six participants were confirmed for each group discussion but

one participant dropped out in each of the first two group discussions while all six confirmed

participants attended the third and final group discussion. Sri Lankan participants also comprised a

similar broad age range and they too were personally known to the researcher. Sixteen

participants were used for the three focus groups and they too were unfamiliar with the research.

The first group discussion consisted of six participants while the other two group discussions

consisted of five participants each as one participant from each of these two groups dropped out.

Table 4.1 shows the general characteristics of the participants.

United Kingdom Sri Lanka Gender Age Occupation Gender Age Occupation

Group 1

Male 35 Employed

Group 1

Male 55 Academic Female 37 Manager Female 53 Housewife

Male 48 Senior Manager Female 53 Director Female 53 Administrator Female 57 Housewife Female 27 Student Female 26 Junior Manager

Group

2

Female 55 Retired Male 26 Manager Male 60 Director

Group

2

Female 29 Housewife Male 31 Unemployed Male 31 Senior Manager

Female 24 Student Male 24 Junior Executive Female 52 Academic Male 26 Manager

Group 3

Male 38 Accountant Male 27 Manager Female 25 Student

Group

3

Female 33 Lawyer Female 50 Administrator Male 34 Administrator

Male 44 Senior Academic Male 28 Student Male 32 Academic Male 28 Administrator Male 26 Student Female 23 Student

Table 4.1: Characteristics of Participants at the time of data collection (April to August 2015)

85

4.4.3 Ethical Considerations

Ethics approval was obtained from the supervisory team as well as Bournemouth University’s

ethics committee prior to approaching potential participants and prior to conducting the six focus

group discussions.

Informed consent was obtained from all the focus group participants prior to the discussions in

line with the guidelines of the Helsinki agreement (World Medical Association 2004). A participant

information sheet (Appendix B) was shared with each participant prior to the discussions and their

informed consent was obtained through a consent form (Appendix C) at the beginning or end of

each group discussion. The information sheet briefly outlined the purpose of the research and

highlighted the participants’ right to withdraw from the research at any time without any negative

consequences. They were also informed through the information sheet that the discussions would

be audio recorded but verbal consent from participants in all six groups was again obtained before

commencing audio recording.

Anonymity and confidentiality were maintained throughout the entire process to ensure that none

of the participants were identified. Participants were identified only by their gender and age in the

presentation of the findings. Given the insensitive nature of the discussions, no emotional harm

was caused to the participants. Further, no participant had to endure any physical duress. All

recordings were safely stored in a private computer only accessible to the researcher and they will

not be made available to any external parties under any circumstances. Transcripts were also

stored in a similarly secure, password-protected manner.

Although not a covert research study by design, the participants were not fully aware of the full

intent of the discussions. Although it was mentioned in the information sheet that the discussion

would revolve around grocery shopping, customer loyalty and loyalty programmes; the specific

research questions were not disclosed. This ensured that participants would not rehearse their

responses and instead the discussions were spontaneous; emerging through group dynamics. Even

though a covert element could be deemed to be present, this was not considered an ethical issue

given the insensitive nature of the discussion topics and also because there was no deception

involved in this research.

86

4.4.4 Data Analysis

The analysis process began with the full verbatim transcription of the audio recorded group

discussions. The transcription of each discussion was done without much delay to ensure that

details were accurately captured (van Teijlingen and Pitchforth 2006). Where applicable, emotions

of the participants, hesitations, reactions and intonations were also included in the transcripts.

Thematic analysis was used to analyse the focus group data. This technique can be defined as a

way of seeing, making sense of and analysing that allows qualitative data to be processed,

analysed and interpreted (Boyatzis 1998). This also allows the researcher to identify, analyse and

report patterns (themes) within the data (Braun and Clarke 2006). Thematic analysis involves the

systematic identification and development of codes and themes interpreted from the data.

Before deciding between manual analysis and using a software package such as NVivo, the

researcher attended two full-day workshops on NVivo to become familiar with the features and

operation of the software. The use of such a software package appeared to be beneficial mainly, in

terms of reducing the need for physical space involving stationery and concerns regarding the

storage and security of transcripts. However, the researcher also identified that such software only

help organise and manage data but do not identify codes and themes from the data. Similarly,

such software also do not interpret findings or draw conclusions from the data. The researcher felt

more comfortable analysing the data manually as it was a way to “get away” from the computer

screen given the ubiquitous usage of a computer in the contemporary research process. This

change in setting allowed the researcher to approach the analysis with a fresh mind-set and

achieve true immersion in the data.

The six phases in the process of thematic analysis as recommended by Braun and Clarke (2006)

were followed to ensure that the analysis was carried out in a systematic manner.

I. Phase 1: Commenced with the verbatim transcription of the focus group discussions which

was followed by repetitious reading while noting down initial ideas using a pencil

II. Phase 2: Involved generating initial codes where interesting features in the data were

systematically highlighted using different colours across the entire data set while a red pen

was used to make note of further thoughts and preliminary interpretations. Data extracts

relevant to each code were collated in a word document which allowed the researcher to

easily access and find these extracts later when writing up the qualitative findings

87

III. Phase 3: Involved collating similar codes into potential themes and relevant data extracts

were subsumed under these themes

IV. Phase 4: These themes were then reviewed to ascertain their relevance and ability to

represent previously identified codes and data extracts

V. Phase 5: The selected themes were named

VI. Phase 6: Involved the presentation of the findings using vivid and compelling extracts from

the data

Given the lack of previous knowledge, the analysis was mainly inductive in nature where the data

drove the analysis. Theoretical analysis was done during the preliminary interpretation of the

findings where cultural variables and structural factors identified in the literature review were

applied to these findings. The themes identified were semantic in nature (Boyatzis 1998) as they

were, at first, taken at surface level after which their broader meanings and implications were

interpreted through the application of previous theory (Braun and Clarke 2006).

4.5 Empirical Study 2: Quantitative Surveys

The preceding qualitative phase explored the influence of cultural and structural differences on

customers’ grocery shopping habits, loyalty and perceptions of loyalty programmes; providing a

preliminary understanding of the role of cultural and structural factors. However, the small sample

size and composition in the qualitative phase meant that the findings were not generalisable.

Therefore, it was necessary to draw a large sample from the two countries to fully examine the

findings. Further, a quantitative approach was required to statistically test the refined conceptual

model developed through the incorporation of the qualitative findings and to make a comparison

of elements within this model in relation to the two countries.

The aims of this quantitative phase were as follows;

I. Ascertain the generalisability of the qualitative findings

II. Measure similarities and differences between the two countries

88

4.5.1 Designing the Questionnaire

The sequential exploratory research design (Qualitative-Quantitative) meant that the

questionnaire was designed to measure the variables identified through the qualitative findings.

The majority of the questions used to measure each of the variables were based on the qualitative

findings. Only the questions used to measure True Loyalty (behavioural and attitudinal loyalty)

were based on existing literature as scales with high reliability and validity were available.

The most pertinent findings from the preceding qualitative phase to be further examined were

identified and the layout of the questionnaire (Appendix D) was moulded to accommodate the

relevant questions. A negative correlation has been reported between the length of a

questionnaire and the response rate where longer questionnaires have been reported to result in

lower rates of response (Lund and Gram 1998; Nakash et al. 2006; Sahlqvist et al. 2011).

Furthermore, previous findings suggest that the response rate could decline by 7.8% when the

length of the survey exceeds four pages (Yammarino et al. 1991). Considering these findings, the

length of the questionnaire was limited to four A4 pages and was designed to fit into one A3 paper

printed double-sided and folded along the long edge to create an A4 booklet. This ensured ease of

use for respondents as well as a cost-effective method.

A combination of vertical and horizontal layouts was used to ensure that the questionnaire was

engaging and to maintain respondents’ interest and engagement with the questionnaire. The use

of open-ended questions in some parts of the questionnaire further improved the overall layout

and enabled greater depth of response where necessary. The overall presentation of the

questionnaire was checked through a sample print run and deemed optimal prior to being

administered.

The majority of the questionnaire consisted of close-ended questions which allowed respondents

to identify the most suitable answer to each question and respond without taking too much time

(Oppenheim 1992). These questions were pre-coded to allow the analysis of the data to be more

efficient and less time consuming (Simmons 2008). While close-ended questions restrict the range

of responses, open ended questions can elicit more in-depth responses as respondents have

greater freedom as they are not bound by a set of predetermined responses (Oppenheim 1992).

The predominant use of close-ended questions was not deemed to be a limitation given the aims

of the questionnaire and the fact that the questions and response groups were informed by the

89

previous inductive phase. Furthermore, no issues related to response choices were identified

during the pre-testing of the questionnaire which is discussed in detail later.

Likert scales with monotonic statements (Likert 1932) and importance rating scales were used for

the close-ended questions. Likert scales are used to gauge a respondent’s extent of agreement or

disagreement with a statement (Oppenheim 1992). There is a debate among scholars regarding

the optimum number of points on a Likert type scale (Croasmun and Ostrom 2011; Leung 2011)

and findings are contradictory and inconclusive. Oppenheim (1992) and Simmons (2008)

recommend the use of a five-point Likert scale but it is claimed that both validity and reliability are

unaffected by the number of scale points (Jacoby and Matell 1971). However, researchers

continue to strive to find the answer to this question and studies propose points ranging from four

(Chang 1994) to as high as eleven (Leung 2011) or even twenty-one (Pearse 2011).

The main consideration in this choice was the inclusion or exclusion of a midpoint. Six-point scales

were used without a midpoint to ensure that respondents answered each question meaningfully,

following the guidance of Pearse (2011). Furthermore, the exclusion of a midpoint encouraged

respondents to be more thoughtful and provide more precise ratings (Adelson and McCoach

2010). The lack of neutrality was also important in obtaining more discriminatory responses given

the comparative nature of the study. However, removing the midpoint forces a respondent to

choose a positive or negative stance instead of remaining neutral (Allen and Seaman 2007) but this

was not deemed to be problematic given the insensitive nature of the research topic. The debate

surrounding this topic is never-ending (Leung 2011) and the choice is generally based on individual

researcher preference (Garland 1991). The six response points were; Agree Fully, Agree,

Somewhat Agree, Somewhat Disagree, Disagree and Disagree Fully.

An importance rating scale was used to measure one question consisting 24 items and a seven-

point scale was used; 7 for Very Important to 1 for Unimportant. This was done to ensure greater

differentiation of the individual items included in the question.

90

4.5.2 Questionnaire Content

Choice of Scales

Overall, scale development was guided by the previous qualitative findings to ensure that the

constructs tested were representative of customer perceptions, grounded in their own vocabulary

(Creswell and Plano Clark 2011). This approach has been argued to ensure greater reliability and

validity of the scales (Creswell 2013). Further, use of focus group findings to refine and develop

measurement scales has been found to enhance respondent comprehension (O’Brien 1993;

Morgan 1997). From a practical perspective, previous scales measuring many of the constructs

were not available at the time of questionnaire development. Therefore, it was decided to develop

the majority of the measurement scales based on the qualitative findings with reference to some

existing scales.

Section A: Grocery Shopping Habits

The first section of the questionnaire (questions 1 to 5) obtained responses pertaining to the

shopping habits of the respondents.

Question 1 identified a respondent’s main supermarket

Question 2 focused on the frequency of purchase at the main supermarket

Question 3 identified the number of supermarkets within easy access from home

Question 4 obtained data on supermarket usage

Question 5 was open-ended and complemented question 4

Question 6 contained 21 items relating to four constructs as identified from the qualitative phase;

Special Occasions, Aversion to Supermarket Own Labels, Captive Loyalty and Aversion to Fresh

Produce in Supermarkets (Specialist Stores). Scales for Special Occasions, Captive Loyalty and

Aversion to Fresh Produce in Supermarkets were not available from the existing literature.

Therefore, scales were developed based on the qualitative findings. Some scales were available for

Aversion to Supermarket Own Labels and these were used to guide the development of two

questions. Previous scales measuring Perceived Quality (Burton et al. 1998; DelVecchio 2001; Bao

et al. 2011) were used to develop the two questions; I am satisfied with the quality of own label

products and I am sceptical about the quality of own label products. The other four questions were

based on the qualitative findings and mirrored participants’ comments such as disliking

supermarkets pushing own labels over national brands and avoiding such supermarkets.

91

Section B: Selecting a Supermarket

Question 7 was comprised of 24 items representing Store Image. These items were extracted from

the literature as well the qualitative findings (Appendix E). Question 8 focused on the importance

of Customer Service. Previous research (Donthu and Yoo 1998; Kueh and Voon 2007; Dash et al.

2009) has used the five SERVQUAL dimensions developed by Parasuraman et al. (1985). However,

these dimensions could not be included given the length restrictions of the questionnaire. Further,

the five questions developed mirrored participant comments on the importance of customer

service. Specifically, these questions were developed to reflect aspects such as disliking a

supermarket and avoiding a supermarket following poor customer service as well as positive and

negative word of mouth.

Section C: Attitudes towards Supermarket Loyalty Programmes

This section included one question with 23 statements developed to measure five constructs;

Attitudes towards Loyalty Programme Tracking, Preference for Instant/Segregated or

Delayed/Aggregated Rewards, Attitudes towards Preferential Treatment, Attitudes towards Soft

Benefits and Attitudes towards Tiered Loyalty Schemes. Since previous scales directly measuring

these constructs were not found, scales were developed based on the qualitative findings;

mirroring participants’ comments.

Section D: About your Main Supermarket

This section contained 14 statements intended to measure three constructs pertaining to

customer loyalty. True Loyalty (behavioural and attitudinal loyalty) was measured through 10

statements adopted from the literature (Appendix F). Given the unavailability of scales measuring

Latent Loyalty and Spurious Loyalty, 4 statements were developed from the qualitative findings to

measure both constructs. These were understood well during the pre-test process. The pre-test

participants understood that “I prefer a different supermarket to my main supermarket” for

instance, denoted a lack of preference for their main supermarket and a greater preference for

another.

Section E: About You

This final section was used to gather respondents’ demographic information. Questions seeking

information on gender and age were the same for the two countries. However, the other

demographic information was obtained using country specific questions. For instance, the income

brackets in the two countries reflected the local currency and appropriate income brackets. This

92

enabled the most effective discrimination between groups. Similarly, level of education was

customised to be country specific and ethnic origin was also customised to reflect the ethnic

groups in each country.

4.5.3 Pre-test and Translation

Thorough pre-testing was undertaken in each country to ascertain whether the target population

comprehends the questions and response options provided as intended by the researcher and is

able to answer meaningfully (Perneger et al. 2015). Pre-testing also aided the proper wording and

sequencing of the questions which can influence the rate of response (Oppenheim 1992).

The primary version of the questionnaire was developed in English and was pre-tested in the UK in

three stages to ensure the clarity of the content, proper comprehension and the appropriateness

of the response options provided. Stage one included pre-tests with ten individuals who

completed the questionnaire in the presence of the researcher and were requested to raise any

queries or concerns related to the questions and response choices. One major change was made

to the questionnaire following this stage; Trust was removed from the questionnaire as individuals

identified that they would respond to the questions used to measure this construct without

adequate discernment. Special Occasions was added to replace Trust and minor changes were

made to the questionnaire by way of wording and phrasing. Two similar stages of pre-testing were

conducted with five individuals in each stage. These individuals were able to respond to the

questions measuring Special Occasions with greater discernment. No new issues were raised and

the researcher was confident that the changes had fully addressed the aims of the pre-test.

The pre-tested English questionnaire was first customised to cater to English speaking respondents

in Sri Lanka. Slight changes were made to localise the wording that did not affect the overall

content of the questionnaire. This version was translated to Sinhala and Tamil; the two national

languages in Sri Lanka. The complexity surrounding translation of a questionnaire from one

language to another has been described as “entering a series of minefields” (Oppenheim 1992,

p.48). Previous scholars (Guillemin et al. 1993; Usunier and Lee 2013) have advised attention to be

paid to the following aspects when translating a questionnaire and care was taken when

developing the original English questionnaire to avoid such issues;

93

Lexical or conceptual meaning- The precise meaning of individual words and the possibility

of multiple meanings/conceptualisations

Semantic equivalence- The possibility of grammatical difficulties when translating

Idiomatic meaning- Avoiding colloquialisms and idioms that may be unique to one

language

Experiential equivalence- Experiences in one country that may not be so common in

another

Usunier and Lee (2013) identified four methods of translation; direct translation, back translation,

parallel translation and mixed techniques while Beaton et al. (2000) suggest a lengthier process

which involves the use of an expert panel.

Translation was carried out by four professional translators; two for each language. The Sinhala

version was done by a translator and back translated by another translator. The source

questionnaire and the target questionnaire did not have any differences. The translated version

was pre-tested with five individuals at different comprehension levels; level of education ranging

from up to GCSE equivalent to postgraduate. No issues were identified and only slight changes

were made following these pre-tests. This ensured comprehension of the questions and response

options as intended by the researcher. The Tamil version was translated by a bilingual translator;

native Tamil speaker, fluent in English. This was back translated by another translator who

identified several issues. These issues were communicated to the first translator and changes were

made accordingly. The revised version was then tested in detail (lasting close to 90 minutes) with a

bilingual person; native Tamil speaker, fluent in English where slight issues were identified. The

process was repeated where the first translator made changes accordingly. The new version was

then pre-tested with five native Tamil speakers at different comprehension levels as with the

Sinhala version. No concerns were raised regarding the comprehensibility of the questions or

response options.

4.5.4 Questionnaire Type

To generalise the findings from the previous qualitative phase, it was necessary to obtain

responses from a sample that was as representative as possible of the population in each country.

For this purpose, a nationwide sample was desired. Self-administered postal questionnaires were

used in the UK due to their relative cost effectiveness in relation to interviewer-administered

94

questionnaires. These questionnaires are also relatively quick to administer and more convenient

for respondents (Bryman and Bell 2011). This type of questionnaire also eliminates the effect of

researcher bias; thus, improving the robustness of the data (Bryman and Bell 2011; Saunders et al.

2012). However, postal questionnaires can have low response rates, usually not more than 20%;

resulting in possible non-response bias (Simmons 2008). Lower response rates (12%-14%) have

been reported in previous PhD theses (Bray 2011; Pyke 2017). Non-response bias presents the risk

of the survey estimate being systematically different from the population's true value (Barclay et

al. 2002); those who do not respond may differ from those who do (Bryman and Bell 2011). The

inability to probe was also identified as a limitation of this method as the absence of the

interviewer meant that there was no way to ascertain whether respondents understood the

questions, response options and instructions. However, the extensive pre-testing of the

questionnaire allayed such concerns. Despite these limitations, this method was most suitable as it

ruled out interviewer bias (inherent with methods such as shopping centre intercept) while

ensuring a random, nationwide sample. Additionally, this method did not present the risk of

selection bias as would have been the case with email databases provided by various research

companies. Further, electronic methods are not as accessible as they exclude those who do not

have access to the internet.

Postal questionnaires could not be used in Sri Lanka due to the unavailability of a nationwide

postal address database. The store intercept technique was ruled out due to permission not being

granted by grocery retailers. Telephone interviewing was also ruled out given the lack of a

sampling frame. Therefore, a mixed-mode survey approach; online questionnaires and hand

delivered questionnaires was used. Online questionnaires were required in addition to hand

delivered questionnaires as reaching geographically dispersed respondents with the latter was not

logistically possible. This method reduced the incidence of coverage error (Dillman et al. 2014)

where online questionnaires provided respondents the opportunity to respond when not covered

by the hand delivered method (Klausch et al. 2017).

Furthermore, this mixed-mode approach improved response times; particularly with online

questionnaires enjoying very low response times. Mixed-mode surveys also have the advantage of

improved response rates, attributable to lower coverage error for instance (Klausch et al. 2017).

Improved response rates have been found to lower non-response bias; leading to lower survey

error (Sala and Lynn 2009). However, reduced non-response error can only be confirmed if

important demographic characteristics of a sample obtained through one method are different to

95

those obtained through a different method (Sala and Lynn 2009). However, it should be noted that

mixed-mode surveys carry the risk of mode effects where responses are influenced by the mode of

survey administration (Lugtig et al. 2011).

Sri Lanka’s low internet penetration rate of 32.1% (The World Factbook 2018b) was previously

identified. However, online questionnaires were a suitable method considering the profile of

organised grocery retail customers in the country. Given the prevalence of organised retailing in

more urban areas with more affluent customers, online questionnaires were aimed at

respondents with greater access to the internet. Despite the different questionnaire

administration methods used in the two countries, it should be noted that the degree of bias is

rather limited. This is due to common advantages and disadvantages shared by these methods.

For instance, mail surveys used in the UK and online questionnaires used in Sri Lanka both can

include a diverse range of questions, obtain moderate amounts of data and are not influenced by

interviewer bias (Malhotra and Birks 2005; Saunders et al. 2012). In a similar vein, these two

methods suffer from low response rates and the inability to probe or clarify (Malhotra and Birks

2005; Saunders et al. 2012).

4.5.5 Sampling Approach and Administration

4.5.5.1 United Kingdom

The questionnaire was administered to a random sample of addresses extracted from the Royal

Mail Postcode Finder. The Postcode Finder contains over 28 million addresses and is the most

comprehensive and up-to-date database of postal addresses (Royal Mail 2016). The alternative

database was the edited electoral register but it was deemed that this database would not be as

representative of the general population as some electors would have opted out of this register

(Sahlqvist et al. 2011). Given the size of the Postal Address Finder database, extracting a random

sample from the entire address list was not straightforward; thus, requiring a systematic

approach. 2,981 postcode districts denoted by the first group of numbers and letters in the

postcode (e.g. BH12) were used as the initial sampling frame and a sample of 100 such postcode

districts was randomly extracted. The 100 postcode districts were placed on a map of the UK to

ascertain whether a broad coverage of the territory was achieved. Figure 4.3 illustrates the areas

covered by these 100 postcode districts. All residential addresses subsumed under these 100

postcode districts were then extracted and a weighted sample of 100,000 addresses was obtained

96

from this database. 3,000 addresses were randomly selected from this list using the random value

allocation and sorting functions found in Microsoft Excel.

Figure 4.3: Survey Coverage of the United Kingdom (Compiled using Google Maps)

Despite being the most comprehensive and up-to-date database for postal addresses (Royal Mail

2016), the Postcode Finder’s weakness compared to the electoral register is its exclusion of

residents’ names; thus, disallowing personalisation (Sahlqvist et al. 2011). Previous research has

highlighted the positive impact of personalisation on response rates (Dillman et al. 2007; Sahlqvist

et al. 2011; Levy et al. 2012) as well as negligible (Gendall 2005) and negative impacts (Houston

and Jefferson 1975). Obtaining a database with residents’ names presented an additional element

of cost and the possible risk of the database being out-of-date. Further, such databases may have

been used previously, resulting in reduced response rates from frequently surveyed addresses.

Similarly, such databases may introduce additional biases given their method of compilation

97

(Dillman et al. 2007). Most importantly, naming could have resulted in the wrong person

responding to the questionnaire. This is because responses were required from the main grocery

shopper in a household and there was no way to identify these persons.

Window envelopes were used to post the questionnaires with the address details displayed on

accompanying cover letters. Cover letters (Appendix G) were printed on Bournemouth University

letterheads and explained the purpose of the questionnaire as well as the importance of response.

The researcher’s contact details were also mentioned so that recipients were able get in touch if

they wished to discuss the research or required clarifications. A pre-paid business reply envelope

was also enclosed in the window envelope. The use of business reply envelopes has been

questioned due to the possibility of respondents getting the impression that it is official or

commercial and result in a lower response rate (Oppenheim 1992). Despite suggestions that

stamped return envelopes could appear more personal and encourage recipients to complete and

return the questionnaire (Harrison et al. 2002), previous findings suggest that the use of stamped

return envelopes over business reply envelopes do not significantly improve the response rate

(Lavelle et al. 2008). Given the cost limitations, business reply envelopes were far more cost

efficient as postage was only payable for envelopes returned; thus, reducing the cost of this

research stage by approximately £1,300.

Although monetary incentives were considered, they were not offered as previous findings

suggest that such incentives do not significantly improve response rates (Nakash et al. 2006;

Cottrell et al. 2015). Furthermore, previous research also suggests that incentives could lead to

response bias as they strongly influence respondent self-selection (Berzelak et al. 2015). While

reminders are recommended to improve response rates (Nakash et al. 2006; Sahlqvist et al. 2011),

a decision needed to be made between sending out questionnaires in smaller batches followed by

reminders and a one-off large batch. Cost and time limitations had to be considered and due to

the substantial cost attached to reminders (Christensen et al. 2015); a one-off batch was posted.

While acknowledging the possible adverse impact on the response rate, it should be noted that

funding to cover the additional cost of reminders could not be secured. The cost effectiveness of

reminders has also been questioned; Christensen et al. (2015) and is further confirmed by a

previous PhD study; (Bray 2011). This study reports a cost of £3.90 for each response without

reminders and £8.09 for those returned following reminders. Furthermore, reminders can be

considered harassing and coercive (Nakash et al. 2006; Schirmer 2009). This would have

threatened the ethical requirement of survey participation needing to be voluntary.

98

4.5.5.2 Sri Lanka

Unlike in the UK, a probability sampling technique was not possible in Sri Lanka as a sampling

frame such as the Postcode Finder or the electoral register was not available. An initial plan was

made to select 20 random cities/towns in the country with at least one supermarket chain in

operation. The researcher was to approach potential respondents on the street at random with a

target of 15 to 20 respondents per area. However, when this method was piloted, it yielded very

poor results for two reasons. Firstly, the adverse weather prevalent during the data collection

period proved to be a deterrent (inter-monsoon rain leading to national flooding). Secondly, the

researcher visited two of the selected areas but was unable to efficiently recruit eligible

respondents; those who use supermarkets in general for grocery shopping. The majority of the

people approached were either those who did not shop in supermarkets or those who only used

supermarkets for topping up. Consequently, this method was not followed and the researcher

attempted to use a store-intercept technique using the premises of the two main supermarket

chains in the country; Cargills and Keells. However, both supermarkets refused to grant permission

to use their premises (including car parks) and approach their customers. An alternative non-

probability sampling method was therefore required.

Given the lowest level of credibility attached to convenience sampling (Saunders et al. 2012), this

method was not used. Further, convenience sampling was not appropriate due to the emerging

nature of organised grocery retailing in the country which meant that a convenience sample would

not have effectively captured respondents who generally shop in supermarkets. Although

considered to be more representative of the population relative to other non-probability methods

(Bryman and Bell 2011), quota sampling could not be used due to the need to obtain responses

from different areas in the country and the lack of access to a large pool of potential respondents

in each area directly known to the researcher. Consequently, a combination of purposive sampling

and snowball sampling was used. Data was gathered initially from respondents directly known to

the researcher and potential respondents had to meet two criteria. Eligible respondents had to be

the main grocery shopper in their household and they had to use supermarkets for more than 50%

of their grocery requirements. Given that the required sample size could not be achieved through

this purposive sample and geographical restrictions of this method, snowball sampling allowed

more respondents and those in different parts of the country to be reached. The criteria

mentioned above also applied to potential respondents captured under this method.

99

The questionnaire was administered in two forms; paper and online. Paper questionnaires were

used to target respondents who did not have access to the internet or did not use emails

frequently. However, this format limited the researcher’s ability to reach geographically dispersed

respondents. Online questionnaires were therefore used to target respondents directly known to

the researcher as well as those who were not known to the researcher; friends, relatives and

colleagues of those known to the researcher. This format allowed geographically dispersed

respondents to be reached. The online questionnaire was designed using the online survey tool

Survey Monkey and was administered via email and social media platforms.

Figure 4.4 illustrates the geographic coverage of the country. The overrepresentation of the

Western province reflects the coverage of organised retailing in the country where such formats

are predominant in the Western province (LMD 2017). Particularly, the low presence of the other

grocery retail chains (excluding the market leader; Cargills) outside of the Western province is

reflected in the sample.

Figure 4.4: Survey Coverage of Sri Lanka (Compiled using Google Maps)

100

4.5.5.3 Sample Size and Response Rates

The required sample size in each country depended on a number of factors including the

method(s) of analysis which are detailed in the next section. Exploratory Factor Analysis (EFA),

Confirmatory Factor Analysis (CFA) and group comparisons were identified as the suitable

statistical analysis methods for this study. Although a universally agreed sample size does not exist

when using the abovementioned analysis methods, a sample size of 300 was chosen for each

country as recommended by Hair et al. (2010) and Field (2013). This was an adequate sample size

for intra-country tests and the combined sample of 600 for inter-country comparisons ensured a

large enough sample.

2,991 questionnaires were mailed in the UK considering the response rates of 12-14% reported in

previous research at Bournemouth University; Bray (2011) and Pyke (2017). Although 3,000

questionnaires were planned; only 2,991 were sent out in July 2016 in the UK. This is due to a

minor shortage of addressed cover letters and pre-paid return envelopes. 40 questionnaires were

not delivered due to various reasons; wrong address, no inhabitants, address does not exist, etc.

resulting in 2,951 being delivered to the random addresses. A total of 302 questionnaires were

returned; a response rate of 10.2% which is below the previous rates of 12-14%. Whilst reminders

were used in the two previous studies (Bray 2011; Pyke 2017), the absence of reminders may have

led to this lower response rate. Of the 302 responses, 279 were useable as some questionnaires

were not adequately filled. Although less than the required sample of 300, this number was

accepted as deviance was less than 10%.

In Sri Lanka, 200 personal messages were sent to potential respondents through a well-used social

media platform. These respondents were asked some screening questions to ascertain their

suitability. They needed to be the main grocery shopper in their household and use modern

grocery retailers for more than 50% of their grocery needs. 30 potential respondents were

reached via email and they were also requested to forward the link to the online questionnaire to

others who met the eligibility criteria which was clearly mentioned in the body of the email. The

number of forwarded emails is unknown which is a characteristic of using an online snowball

sample. While the response rate is also unknown, 122 responses were obtained with 117

responses being useable. 169 paper questionnaires were returned of which 161 were useable.

Approximately 250 paper questionnaires were distributed resulting in a response rate of 67.6%. A

total of 291 responses were obtained in Sri Lanka out of which 278 were usable. 593 responses

were obtained from the two countries in total.

101

United Kingdom Format Total (Delivered) Total Returned Response Rate Useable

Postal Questionnaires 2,951 302 10.2% 92.4% Sri Lanka

Format Total Total Returned Response Rate Useable Paper Questionnaires 250 (Approximately) 169 67.6% 95.9% Online Questionnaires Unknown 122 N/A 95.3% Table 4.2: Response Rates

4.5.6 Data Analysis

4.5.6.1 Data Entering, Cleaning and Imputation

Data were initially entered into Excel sheets which made the cleaning of the data easier due to the

various formatting options available. Responses were input into Excel by the researcher as

responses were obtained which enabled the researcher to gain a preliminary understanding of the

patterns in the data from the two countries.

The data cleaning process began with the search for cases with severe missing value issues. The

=COUNTBLANK formula in Excel was used to identify the number of missing values per case. The

data included in this analysis were those captured by Likert type scales since these variables were

to be subjected to a factor analysis in which missing values could adversely affect the analysis.

Following the guidance of Gaskin (2017a), cases with more than 10% missing values for Likert type

scale data were removed from the samples. Consequently, 5 cases were removed from the 279

cases in the UK and 2 cases from the 278 in Sri Lanka.

Prior to imputing missing values (for the 87 Likert type scale items), the severity of the missing

values was established. Little’s (1988) MCAR test for Values Missing Completely At Random was

significant in both countries which meant that cases did not have to be deleted listwise or pairwise

(Garson 2015). Another test (MAR) was conducted to ascertain whether data were Missing At

Random. Since none of these 87 items in the two countries had cases with more than 5% missing

values (Garson 2015), advanced imputation techniques such as multiple imputation or missing

values analysis were not required. Consequently, simple imputation was done for these 87 items

with missing values where the median of all nearby points was used. The median was used instead

of the mean as recommended in the case with Likert type scale data which are categorised as

ordinal data (Jamieson 2004).

102

Skewness and Kurtosis of the data for the 87 Likert type scale items from the two countries were

then checked using SPSS version 22 to ascertain the normality of data distribution and no issues

were identified. The normality of the data is an assumption made prior to undertaking statistical

tests such as Exploratory Factor Analysis (Field 2013). Skewness and Kurtosis in some instances

were to be expected given the comparative nature of the study. This is because respondents from

one country would respond to items in contrasting fashion to those from the other. Similarly,

outliers were also checked in relation to categorical data and no issues were identified.

The level of engagement was also checked for the 87 Likert type scale items, again using Excel. The

=STDEV formula allowed the standard deviation of each case to be checked. While the level of

engagement with the 63 6-point Likert type scale items was satisfactory; greater than 0.5 (Gaskin

2013) in both countries, some issues were identified in relation to the 24 7-point rating scale

items. This issue was more severe with the Sri Lankan data as a higher number of cases were

identified to be “poorly engaged” in comparison to the UK data. Steps taken to address this issue

will be discussed later when the respective analysis is discussed. The cleaned data were then

inputted into SPSS version 22.

4.5.6.2 Exploratory Factor Analysis

Exploratory Factor Analysis hereinafter referred to as EFA is a data reduction technique which

allows data from a large set of variables to be reduced or summarised using a smaller set of

factors (Pallant 2016). The EFA was conducted on the 63 items employing 6-point Likert type

scales to measure the various constructs identified from the preceding qualitative phase. An EFA

was required firstly to ascertain whether variables intended to measure a particular construct

would be subsumed under a factor representing that construct. Secondly, an EFA was required as

53 of the 63 items used were new items which were not based on existing theory but on findings

from the preceding qualitative phase.

Maximum Likelihood was the method used when running the EFAs as the pattern matrices were to

be used in subsequent confirmatory factor analyses (Gaskin 2017b). Oblique Rotation (Promax)

was used given the assumption that the constructs are not unrelated as assumed when using

Orthogonal Rotation techniques (Hair et al. 2010; Field 2013). Prior to interpretation, the

appropriateness of using an EFA was ascertained by exploring the Kaiser-Meyer-Olkin Measure of

Sampling Adequacy (KMO) and Bartlett’s Test of Sphericity (Hair et al. 2010) where values greater

than 0.7 are required for an acceptable KMO and the latter test should be significant (Field 2013).

103

When interpreting the pattern matrices, the significance of the loadings was observed. Hair et al.

(2010) provide guidelines for identifying significant factor loadings based on sample size. A loading

of 0.35 is recommended for a sample of 250 and 0.3 for a sample of 300. Given the sample sizes in

the UK (274) and Sri Lanka (276), 0.35 was used as the minimum threshold when determining

significant factor loadings. Further, variables that did not load under any of the factors were

removed as they did not reflect any of the factors. Similarly, communalities of the variables were

also checked when reiterating the analysis where variables with communalities less than 0.3 were

removed (Gaskin 2017b). The deletion of variables with low communalities was done in

consultation with the reliability scores of the variables subsumed under each factor.

With regard to the number of factors to be extracted, two criteria were used. Firstly, an a priori

criterion was applied (Hair et al. 2010). The 63 (6-point Likert type scale) items were used to

measure 12 constructs identified in the preceding qualitative phase. Therefore, the researcher was

aware of the possibility for 12 factors to be extracted. Secondly, Kaiser’s criterion (Field 2013;

Pallant 2016) which suggests that only factors with eigenvalues greater than 1 should be extracted

was applied. While the Scree Test is another criterion (Hair et al. 2010), it was not used as the

points of inflexion were ambiguous and did not suggest clear cut-off points similar to Kaiser’s

criterion. Given that the final pattern matrices were to be used for confirmatory factor analyses, a

minimum of 3 variables with significant factor loadings were required for each factor (Hair et al.

2010). Thus, any factors with less than 3 significantly loading variables were not extracted and the

factor analysis was reiterated until no factors with less than 3 significant loadings were present.

Following the extraction of the final pattern matrix for each country, the reliability and validity of

the factors were tested. Cronbach’s Alpha was used to test the reliability of the factors (Field

2013). A minimum alpha of 0.7 was set as the standard (Nunnally 1978) while variables with

Corrected Inter-Item Correlations below 0.3 were deleted. Similarly, Cronbach’s Alpha if Item

Deleted was checked for each variable where any variable whose corresponding value was greater

than the overall alpha was considered for deletion. Such considerations however were governed

by the Corrected Inter-Item Correlation of each variable where any variable with a correlation

below 0.3 was deleted. In the event of such deletions, the exploratory factor analysis was iterated

to ensure that the structure of the pattern matrix was not affected (Field 2013). Convergent

validity of the factors was tested by scanning the loadings of each factor where high loadings

(above 0.7) and the average loading being 0.7 determined convergent validity (Gaskin 2017b).

104

Divergent validity was ascertained by the absence of cross-loadings and the factor correlation

matrix not having correlations greater than 0.7 (Gaskin 2017b).

Upon establishment of the reliability and validity of the extracted factors in each country, the

pattern matrices were used in the ensuing confirmatory factor analysis.

4.5.6.3 Confirmatory Factor Analysis

SPSS AMOS version 22 was used to run the confirmatory factor analyses using the final pattern

matrices. Confirmatory Factor Analysis, hereinafter referred to as CFA, is a more complex and

sophisticated technique used “to test (confirm) specific hypotheses or theories concerning the

structure underlying a set of variables” (Pallant 2010, p.181). This analysis enabled the further

testing of the robustness of the data in terms of the composite reliability, convergent validity and

divergent validity of the factors (constructs) prior to carrying out subsequent analyses.

The final pattern matrix for each country was inputted into AMOS using the Pattern Matrix Model

Builder plug-in which automatically draws the diagram and transfers corresponding factor scores

into the measurement model. The model fit for each country was checked after an initial analysis

and the guidelines provided by Hair et al. (2010) and Gaskin (2017c) as shown below were

followed. The initial model fit in either country was not satisfactory, thus the modification indices

were checked and large, within factor error terms (> 20) were co-varied which resulted in

improvements in the model fit (Gaskin 2017c).

Measure Threshold Chi-square/df (cmin/df) < 5 or ideally < 3 p-value of the model > .05 (Not applicable with large samples) CFI > .90 or ideally > .95 SRMR < .09 RMSEA Between 0.03 and 0.08 PCLOSE > .05 Measurement Invariance

Measurement invariance ascertains whether survey items are interpreted the same way by

members of different groups (Cheung and Rensvold 2002). Ensuring measurement invariance is of

particular importance when comparing different cultural groups (Chen 2007). If measurement

invariance is not confirmed, comparisons of different cultural groups could be inconclusive and

erroneous (Steenkamp and Baumgartner 1998).

105

Configural invariance and metric invariance tests were run on the full sample model (n= 550) with

two groups; UK and Sri Lanka. Configural invariance was confirmed by the satisfactory model fit

thresholds (CMIN/DF < 3; CFI > 0.9; SRMR < 0.09; RMSEA < 0.05; PCLOSE > 0.05) when the two

groups were estimated freely (without constraints). Metric invariance was tested by constraining

the loadings (regression weights) to be the same across the two countries. The Chi-square

difference (χ2) test between the unconstrained and fully constrained models was non-significant,

which confirmed metric invariance. Therefore, differences in item scores can be meaningfully

compared between the two countries and such differences will reflect cross-national differences

underlying the constructs (Steenkamp and Baumgartner 1998). Given the sample size for each

country (274 in the UK and 276 in Sri Lanka), the results of the invariance tests can be treated with

confidence (Meade and Bauer 2007).

Common Method Bias

The focus then moved towards testing for common method bias which is variance that can be

attributed to the method of measurement rather than to the constructs represented by the

measures (Podsakoff et al. 2003). MacKenzie and Podsakoff (2012) identified a list of potential

causes and these were addressed where possible, prior to questionnaire development and

administration. Issues with verbal ability, education or cognitive sophistication were addressed by

pre-testing the questionnaire with people at different education levels in both countries. Lack of

experience was particularly important in Sri Lanka due to the developing nature of organised

grocery retailing. Therefore, steps were taken to recruit respondents with adequate experience.

Clear statements were used in the scales to avoid item ambiguity and complex or abstract

questions. Similarly, double-barrelled questions and questions requiring recall were avoided.

Written presentation of the questionnaire in both countries avoided the potential issues

surrounding auditory only presentation.

However, common method bias can be caused for example, as a consequence of using a single

data collection method or having similar rating scales for the dependent variable(s) and the

independent variables. Obtaining responses to dependent and independent variables from the

same source further runs the risk of creating common method bias (Podsakoff and Organ 1986;

Chang et al. 2010). Although measures were taken to reduce the possibility of common method

bias occurring, tests needed to be carried out to measure and control for its presence. Podsakoff

et al. (2003) outline several techniques to test for common method bias. The marker variable

106

technique and the common latent factor technique as recommended by Podsakoff et al. (2003),

Richardson et al. (2009) and Gaskin (2017c) were used.

In the marker variable technique, True Loyalty was used as the directly measured latent methods

factor as recommended by Podsakoff et al. (2003) and Gaskin (2017c). The Chi-square difference

(χ2) was significant when the model with the marker variable and the standard model were

compared. The common latent factor technique also provided the same result, where the

difference in χ2 was significant when the unconstrained CMB model and the fully (zero)

constrained CMB model were compared. Therefore, the presence of common method bias was

confirmed.

The presence of Common Method Bias can be explained using the reasons cited by Podsakoff et al.

(2003) as only one method was employed to collect data in the UK; postal questionnaires. Further,

all questions employing 6-point Likert type scales (Chang et al. 2010) and some negatively worded

questions may have contributed to this. Common Method Bias is also explained as the presence of

substantial shared variances between factors (constructs). This too is possible as some constructs;

preferential treatment, soft benefits and tiered schemes are somewhat conceptually similar,

although used in this research as different constructs. Another explanation could be the large

share of the variance in the pattern matrix being accounted for by a small number of factors; 3

factors in both countries contributing to almost 40% of the variance. Although it is noted that

avoiding Common Method Bias in inter-country surveys is near impossible; Chang et al. (2010),

both countries’ measurement models included a Common Latent Factor in the CFA which corrects

the measures for Common Method Bias.

Model fit for each country was again checked after which the composite reliability; above 0.6

(Bagozzi and Yi 1988), construct validity; average variance extracted above 0.5 (Bagozzi and Yi

1988) and divergent validity; square root of the average variance extracted greater than inter-

construct correlations (Fornell and Larcker 1981; Anderson and Gerbing 1988) of the constructs

were tested. Factor scores were then imputed for the two countries with the Common Latent

Factor retained.

107

4.5.6.4 Inter-Country Differences Comparison of Constructs

Once the factor scores were imputed for each country, the analysis moved on to comparing

similarities and differences between the two countries in respect of the constructs (factors). The

two countries’ factor scores were copied into a separate SPSS file for this analysis with a

categorical variable used to distinguish the countries (1; UK and 2; Sri Lanka). Given the non-

parametric nature of the data, Mann-Whitney U Tests (Pallant 2016) were used to test for any

differences. Any difference was accepted as statistically significant if the significance (p value) was

below 0.05. In the case a significant difference was noted, follow up median scores were

calculated to ascertain the direction of the difference. All significant differences were followed up

by a test for effect size using Cohen’s (1988) criteria where .1 represents a small effect, .3 a

medium effect and .5 a large effect. This test was used to identify the magnitude of any significant

differences.

Store Image

Prior to analysis, the level of engagement with this question was checked. Using the threshold of

0.5 (Gaskin 2013), 8 cases from the UK failed to meet this threshold while 39 Sri Lankan cases fell

below this threshold. Subsequently, 266 cases from the UK and 237 cases from Sri Lanka

respectively were used in a Mann-Whitney U Test where significant differences were followed up

by calculations of median scores and effect sizes.

4.5.6.5 Within-Country Differences Loyalty towards Main Supermarket

This analysis focused on the type and degree of loyalty towards the main supermarket. It was

identified from the focus group data that differences in customer loyalty exist within the two

countries.

Given the non-parametric nature of the data and the need to include more than 2 groups in the

analysis, Kruskal Wallis (H) Tests; Pallant (2016) were conducted for both countries separately. It

should be noted however, that while this test demonstrates whether significant differences exist

between the groups, it does not specify where these differences lay. Therefore, when significant

differences were reported, post-hoc Mann-Whitney U Tests were carried out on pairs of groups

108

with Bonferonni corrections; the normally accepted significance level of 0.05 was divided by the

number of paired post-hoc tests. This reduced the risk of rejecting the null hypothesis incorrectly

as the level of significance was more stringent (Field 2013). 11 supermarkets were reported by the

UK respondents as being their main supermarket while 6 were reported by their Sri Lankan

counterparts. The acceptance level of significance was thus dependant on the number of paired

tests required for each country where n represents the number of supermarkets;

United Kingdom: ((n-1)/2) x n); ((11-1)/2) x 11) = 55 paired tests

Sri Lanka: ((n-1)/2) x n); ((6-1)/2) x 6) = 15 paired tests

Consequently, the acceptable level of significance for the UK was 0.00091 (0.05/55) while it was

0.0033 (0.05/15) for Sri Lanka. Such low levels of significance meant that differences would not be

detectable; thus, a decision was made to group supermarkets based on existing industry

categorisations (Mintel 2016c) and participants’ categorisations. In the case of the UK, the Big Four

supermarkets; Tesco, Sainsbury’s, Asda and Morrisons were grouped into one category labelled

“Big Four”. Aldi and Lidl who are known for their value dominant logic were grouped into the

second category labelled “Discounters” while Marks & Spencer and Waitrose were grouped

together and labelled “Premium”. The fourth and final category included The Co-op, Iceland and

other supermarkets and was labelled “Other”. In the case of Sri Lanka, Cargills was taken in

isolation given its position as the clear market leader and comments made by participants in

relation to loyalty. Keells and Arpico were grouped together and the category was labelled

“Better” as they were commonly referred to by participants. The third and final category labelled

“Other” included Laugfs, Sathosa and other supermarkets. Consequently, the required level of

significance was lowered to 0.0083 for the UK (6 paired tests) and 0.017 for Sri Lanka (3 paired

tests). The analysis was then run for each country using the categories and the imputed scores for

the constructs representing True Loyalty and Spurious & Latent Loyalty. Median scores and effect

sizes were calculated when significant differences were observed.

It was also mentioned by some Sri Lankan participants whose main supermarket was either Keells

or Arpico that they felt a lack of alternatives; a sense of Captive Loyalty given the fact that no

better alternatives were available to them. Although no such comments were made by UK

participants, it was intriguing to identify whether the degree of Captive Loyalty varies according to

the main supermarket. Thus, Kruskal Wallis (H) Tests were conducted to ascertain whether

differences in Captive Loyalty could be observed in relation to the main supermarket. A similar

procedure as mentioned above; follow-up Mann Whitney U Tests with Bonferonni corrections if

109

differences are significant, median scores and effect size calculations if paired comparisons are

significantly different was followed.

Captive Loyalty

Q3; Availability of supermarkets within easy access from home was used in this analysis together

with the imputed scores for the construct referring to Captive Loyalty. Although the difference

between Captive Loyalty between the two countries was tested under inter-country differences,

this analysis was used to ascertain whether the availability of supermarkets would correspond

with the level of Captive Loyalty for each respondent. 7 response categories were used for Q3; 0,

1, 2, 3, 4, 5 or more and not sure. Respondents who picked the seventh response category (not

sure) were excluded from the analysis and the remaining 6 categories were grouped into 2

categories; Low Availability (0, 1, and 2) and High Availability (3, 4 and 5 or more). Using these two

categories to represent ‘Availability’ and the Captive Loyalty construct, Mann-Whitney U Tests

were carried out to ascertain whether respondents with low availability reported greater Captive

Loyalty. Any significant differences were followed up by calculations of median scores and effect

sizes.

Supermarket Usage and Level of Loyalty

This analysis focused on the degree of customer loyalty in relation to the monthly usage of

supermarkets (Q4; number of supermarkets used per month). The five response options (1, 2, 3, 4,

and 5 or more) were categorised into two groups; Low Usage (1 and 2) and High Usage (3, 4 and 5

or more). Mann-Whitney U Tests were carried out to ascertain whether differences in True

Loyalty, Spurious & Latent Loyalty and Captive Loyalty could be observed between the two groups

in respect of their usage. Any significant differences were followed up by calculations of median

scores and effect sizes.

Analysis of Open-Ended Questions

Responses to open ended questions were analysed through a content analysis where responses

were ranked based on how frequently they were mentioned. Given that the two open-ended

questions were used to complement their preceding close-ended questions, it was important to

identify the main reasons identified by respondents in support of their responses to the preceding

fixed response options. The less qualitative content analysis was used purely to identify patterns

so that frequencies could be identified. The intention was not to present an in-depth account.

110

Furthermore, the relatively short responses did not allow for a more qualitative analysis technique

to be applied.

Non-Usage of Path Analysis or Multiple Regression

Although construct scores were computed following the Confirmatory Factor Analysis in each

country, path analysis or multiple regression analysis was not conducted. Whilst True Loyalty and

Spurious Loyalty & Latent Loyalty were measured in relation to respondents’ respective main

supermarket, the other constructs were not measured in the same manner. Therefore, causal

effects could not be measured. For instance, the effect of Aversion to Supermarket Own Labels,

Expectation of Customer Service or Expectation of Preferential Treatment on True Loyalty could not

be measured because these constructs were measured in a generic manner, not focused on the

main retailer. Therefore, it was not conceptually meaningful to measure causal effects. Further,

the aim of the statistical primary research phase was to measure inter-country and intra-country

differences in respect of the key constructs.

4.6 Chapter Summary

This chapter presents the mixed methods research design employed in this study; justifying the

use of a Pragmatic philosophical stance. A detailed account of the sequential mixed methods

approach is provided, followed by a description of the first primary research phase; focus group

discussions. The rationale for using focus groups is presented followed by a detailed account of the

sampling process, data collection process and data analysis process. The second primary research

phase; questionnaires is similarly detailed, focusing on questionnaire design, content, pre-test

process and translation. The type of questionnaire used and sampling techniques are justified with

references to country specific dynamics. The sample size and response rates are presented for

each country followed by a detailed account of the quantitative data analysis methods utilised in

this study.

111

Chapter 5

Section 1

Preliminary Interpretation of

Qualitative Findings

112

5.1 Introduction

This chapter is divided into two sections and the first section presents the interpretation of the

qualitative findings from the focus group discussions. The second section presents the quantitative

results. The key qualitative findings are used to refine the conceptual framework developed

following the literature review. This refined conceptual framework guided the development of the

questionnaire used in the subsequent quantitative phase. The results and findings from this phase

were used to further investigate the relationships identified in the refined conceptual framework.

5.2 First Empirical Study- Focus Group Discussions

Thematic analysis of the focus group data resulted in the identification of three broad themes

namely; Grocery Shopping Habits, Factors Affecting Store Choice and Customer Loyalty & Loyalty

Programmes. A template of the key findings and sub-findings pertaining to each broad theme is

presented under each section. Although comparisons are made between the two countries based

on the qualitative findings, these are merely speculative; highlighting the content to be further

investigated in the ensuing quantitative survey.

5.2.1 Grocery Shopping Habits

Figure 5.1 shows the key findings and sub-findings. The following colour coding has been used;

black (both countries), red (UK) and blue (Sri Lanka). The possible implications of these findings on

customer loyalty are discussed within this section.

Grocery Shopping Habits Normal Shop Big Shop Top-up Shop Special Occasions Dinner Party Birthday Christmas Treat Own Labels Accepting Own Labels Averse to Own Labels Shopping Tendencies Preference for Specialist Stores Avoid Supermarket Fish & Meat

Patronise Several Stores Figure 5.1: Key Themes and Sub Themes Pertaining to Grocery Shopping Habits

113

In both countries, two general shopping types were identified from the discussions; main shop and

top-up shop. In the UK, participants mentioned using the larger stores of a grocery retailer for

their bigger shop and convenience type stores for topping up. Participants suggested that this

behaviour was due to lower prices and deals at larger stores as opposed to relatively high prices

and lack of deals in convenience type stores. This point was exemplified by one participant who

mentioned;

There are kinds of deals that they don’t do in those kind of... yeah the smaller ones like the

buy one get one frees that they do in those big supermarkets... so it’s good for those one-

off like bread and milk... to do a big shop is a bit cheaper at a supermarket...

(Male, aged 38: UK)

Participants further mentioned that they would use their local corner shop to top-up but it was

identified by them that these corner shops are being replaced by convenience type stores of larger

grocery retailers. In contrasting fashion, Sri Lankan participants mentioned the use of large grocery

retail stores or unified wet markets known as “Arthika Madhyasthana” (Economic Centres) found

in urban areas. Given the lack of convenience type stores, participants mentioned that they would

either top-up at the same grocery retail chain store or at a nearby traditional retail store.

Although not mentioned during any of the Sri Lankan focus group discussions, UK participants

mentioned that their shopping behaviour would be influenced by special occasions such as a

dinner party, birthday or Christmas. Such occasions according to them influence the type of

products purchased and move them away from their usual store(s) to more premium retailers

such as Waitrose and Marks & Spencer;

I have shopped online with Waitrose but that’s only like Christmas when they do all their

nice meat... (Female, aged 52: UK)

It became clear through the group discussions that this tendency was a result of participants

perceiving Waitrose and Marks & Spencer to be relatively superior in terms of product quality and

reputation; possibly resultant of positioning strategies used by these grocery retailers.

Furthermore, occasional shopping in such stores was considered by participants as a treat;

It’s more of a treat for us doing Waitrose or Marks and Spencer’s (Male, aged 35: UK)

114

In contrast, Sri Lankan participants did not mention being influenced by such special occasions.

Moreover, high-end grocery retailers such as Waitrose and Marks & Spencer are not present in Sri

Lanka and such positioning strategies are not used by any grocery retailers. Such behaviour could

have a negative influence on customer loyalty to grocery retailers in the UK. This is because

customers may sway towards these premium retailers for special occasions; reducing behavioural

loyalty to their main retailer. Further, such behaviour questions their attitudinal preference for

their main retailer. As a result, a certain degree of spurious loyalty is possible towards the

mainstream retailers and the discounters in the UK. It is also plausible that customers have a

degree of latent loyalty towards the two premium retailers.

UK participants mentioned that own labels make up a major portion of their shopping baskets.

They were very familiar with own labels and none of the participants expressed any concerns or

dissatisfaction with these own label products. In stark contrast, where such own labels are

relatively new, Sri Lankan participants expressed aversion to such products. Participants expressed

their unhappiness with grocery retail chains pushing their own labels and depriving them of their

preferred brands. Participants even expressed their unwillingness to shop at certain retailers as a

result and this was typified by the following comment;

Like looking for ice cream at Cargills. They used to have all the lot but now it’s just Cargills

ice cream, so I never go there... (Male, aged 26: Sri Lanka)

Negative perceptions of own labels as mentioned by Sri Lankan participants appeared to be driven

by quality concerns and preferences towards established brands with one participant saying;

I have used the dish wash bar; the Keells dish wash bar and the toilet cleaner but my

mother used to complain it’s not good (Male, aged 27: Sri Lanka)

In such contexts, grocery retailers may find it difficult to enjoy much success with their own labels.

As mentioned before, pushing own labels and depriving customers of their preferred brands could

result in the generation of negative attitudes. This could adversely affect their loyalty as negative

attitudes could lead to reduced behaviour.

UK participants, despite the relatively advanced retail structure, mentioned a preference to buy

their fish and meat from specialist stores whenever possible. This was alluded to the higher

perceived quality and relatively greater value for money at such specialist stores in comparison to

supermarkets. Their comments also highlighted a preference to buy other fresh produce from

115

local markets due to the greater perceived quality of products. Apart from such specialist stores,

UK participants also mentioned a preference towards the more high-end grocery retailers such as

Waitrose and Marks & Spencer for fresh produce. This preference was linked to the better quality

perceptions of these two retailers; a result of positioning strategies as previously mentioned. This

was typified by the following statement;

Usually would only get meat… some of the “posher” ones like Waitrose and Marks or

whatever’s got eh, um, counter possibly (Male, aged 35: UK)

Similarly, Sri Lankan participants also made clear their preferences towards specialist stores for

fresh produce driven by their perceptions of such stores providing better value for money, quality

and freshness. Although UK participants mentioned a willingness to shop for fresh produce at

premium retailers such as Waitrose and Marks & Spencer, Sri Lankan participants mentioned that

they would avoid buying fresh produce from supermarkets in general;

Come to think of it, I don’t buy fish from the supermarket (Female, aged 53: Sri Lanka)

This could be a result of perceptual differences related to value for money, freshness and quality

as well as the absence of such premium retailers in Sri Lanka. Although referred to as fresh

produce in general, Sri Lankan participants appeared to be more concerned with the quality of

meat and fish compared to vegetables and fruits. A perceptual difference between two categories

of retailers became clear in this regard. Participants did not express concerns about the quality of

fresh produce in general offered by Keells and Arpico; two retailers they perceive to be better than

the other retailers. On the contrary, they were highly critical of the quality of fresh produce at the

market leader; Cargills as well as Laugfs and Sathosa.

Don’t go to (Cargills) Food City or Laugfs if you want to buy meat or fish… I will never buy

meat from Cargills (Male, aged 55: Sri Lanka)

Cargills and Laugfs... Laugfs in particular, oh my, is very... the meat section there stinks. I

went to their store… once and the beef in the meat section smelt really bad, I felt the beef

was very old... (Female, aged 53: Sri Lanka)

Negative perceptions of grocery retailers and preferences towards specialist stores could affect

customer loyalty negatively in both countries. Customers’ behavioural loyalty could be negatively

influenced by such perceptions as they may shop away from grocery retailers. Preferences for

116

premium retailers in the UK could create latent loyalty towards such retailers and result in

spurious loyalty to customers’ main retailers; a phenomenon similar to special occasions. Keells

and Arpico may enjoy greater loyalty from customers due to favourable perceptions. On the other

hand, Cargills as well as Laugfs and Sathosa may be avoided by customers; especially when buying

fresh meat and fish. Similar to the UK, such preferences and behaviour could lead to greater latent

loyalty to premium Keells and Arpico whilst resulting in spurious loyalty to the other retailers.

Patronising several grocery retailers and shopping around emerged as a common habit from the

UK group discussions whereas Sri Lankan participants did not mention such behaviour. Such

behaviour in the UK could be a result of the vast number of grocery retailers in operation

compared to Sri Lanka; providing customers a wider range of options. This was identified by one

participant who said;

And I guess that’s ‘cos we’ve got a lot of choice seemingly wherever we are in this

country... we’ve got such a choice (Female, aged 52: UK)

As a result, purchasing could be spread across a wider range of retailers in the UK compared to Sri

Lanka and this could have a negative influence on behavioural loyalty towards a customer’s main

retailer. Furthermore, patronising multiple stores could also have a negative influence on true

loyalty; repeat patronage and relative attitude to a customer’s main store.

The key findings are summarised below. Findings in red are exclusive to the UK and those in blue

are exclusive to Sri Lanka. Findings pertaining to both countries are in black whilst the (+) and (-)

symbols denote the possible positive or negative implications for customer loyalty. UK participants

tend to shop away from their main grocery retailer during special occasions; resulting in a lack of

behavioural loyalty towards their main retailers. Greater promiscuity amongst UK participants may

negatively affect their behavioural loyalty to their main store. A preference for specialist stores in

both countries in respect of fresh produce could have a negative impact on the behavioural loyalty

to the main retailer. Sri Lankan participants expressed an aversion to own labels due to their

unfamiliarity and perceptions of low quality compared to national brands. Retailers promoting

such products over national brands were perceived negatively; suggesting the possibility of

negative attitudes towards such retailers.

117

Figure 5.2: Key Findings pertaining to Grocery Shopping Habits

5.2.2 Factors Affecting Store Choice

Each participant identified factors affecting their store choice and factors initially missed by

participants were identified as the discussion progressed. Although individual importance placed

on these factors varied, the participants did discuss and identify factors most important to them.

These factors are shown in Table 5.1 below.

United Kingdom Sri Lanka Price, Deals & Offers Freshness & Quality

Convenience Customer Service

Ethical Considerations Price

Loyalty Programmes & Affiliated Promotions Range & Choice

Quality, Range/Choice and Store Layout Convenience

Table 5.1: The Top Five Factors Affecting Store Choice

Despite participants in all six focus groups mentioning price as being an important factor in

deciding where to shop, UK participants placed greater emphasis on this factor. For them, price

was accompanied by deals and offers while it was price in general for their Sri Lankan

counterparts. Given the greater availability of deals and offers in the UK and the widespread use of

such initiatives by grocery retailers, UK participants were more mindful of such deals and offers.

They also mentioned that they would be attracted to such initiatives that could influence their

shop choice. Further, some participants mentioned their active search for such bargains when

shopping for groceries; “I’m always on the lookout for bargains” (Female, aged 52: UK). As a result

of the focus on these elements, store patronage behaviour could be negatively affected in the UK.

Given the vast choice of grocery retailers in the UK, customers may be prone to switching as

grocery retailers compete on price.

Promiscuity (-)

Specialist Stores (-)

Grocery Shopping Habits

Special Occasions (-)

Own Labels (-) Key to Colour Coding Red- UK Blue- Sri Lanka Black- Both Countries

Customer Loyalty

118

Despite convenience being mentioned by Sri Lankan participants, UK participants seemingly placed

greater emphasis on this factor. The structure of UK grocery retailing with its multitude of

convenience type stores may have driven customers towards convenience. Similarly, the

convenience orientation of customers may have contributed to the growth of the convenience

sector and the rise of convenience type stores. Convenience at times was mentioned to be the

bottom-line with one participant even mentioning that convenience would override price, deals

and offers;

Uh, yeah, convenience to a point, yeah... Won’t go driving looking out for an Asda if there

was no one near, doesn’t matter what the price is, it’s not going to work.

(Male, aged 48: UK)

Ethical considerations were also mentioned by UK participants. Participants were aware of

allegations such as profiteering and exploiting suppliers, levelled against certain grocery retailers

and their holding companies. Participants further suggested that they would like to shop at more

ethical retailers but admitted to being restrained because of cost considerations. The importance

of fair-trade and organic particularly, was identified; their preference for these products and at

times, willingness to pay a reasonable premium for them. A preference towards certain grocery

retailers because of their efforts to promote such products became apparent; “I like Tesco’s ‘cos

they stock the best organic range” (Male, aged 31: UK).

However, participants confessed that they were restricted by the relatively high prices. Finally,

participants mentioned the importance of sourcing with reference to local produce, importing

products and the impact on British farmers. Whilst identifying their preference towards local

(British) produce, participants admitted their guilt associated with purchasing imported products

at the expense of local products. They further expressed their dissatisfaction with grocery retailers

importing goods when they could source such goods locally with one participant saying;

You know it infuriates me that New Zea- Australia… lamb is a by-product of their main

industry which is wool and yet we import it by the ton at the sacrifice of the British

farmer… (Female, aged 53: UK)

In stark contrast, none of the Sri Lankan participants mentioned the importance of such ethical or

provenance considerations. This could be a result of a lack of awareness in Sri Lanka and the lack

of awareness building initiatives undertaken by organisations or the government. Further, the

119

freedom enjoyed by the media in reporting unethical practices of organisations in Sri Lanka is very

low compared to the UK. As a result, customers may not always be informed of such matters as

such news may not be publicised. A good example is the horsemeat scandal in the UK where guilty

grocery retailers such as Tesco, despite their market leadership, were named and shamed.

Some UK participants mentioned that their store choice would be influenced by loyalty

programmes but this was not identified in Sri Lanka. This could be primarily a result of the lifecycle

of these programmes. While these programmes are established and widely used by UK customers,

they are still a novelty in Sri Lankan grocery retailing and customers are still getting used to, and

are relatively unfamiliar with the concept of loyalty schemes. UK participants also mentioned the

impact of grocery retailer issued credit cards and the additional points on offer influencing their

purchasing. The leading grocery retailers such as Tesco and Sainsbury’s have introduced such cards

while the leading supermarkets in Sri Lanka do not have such cards.

Apart from these factors, quality was identified by the UK participants as an important factor

influencing their choice of retailer. Furthermore, some participants mentioned that on special

occasions, they would opt for quality over price;

If it’s for me, I don’t really care, I’m happy... I’ll have… but like if I’m preparing for other

people, I’m like yes! Big old beautiful thing with great presentation so maybe good meat

and good veg... (Male, aged 32: UK)

The greater importance of quality on special occasions goes to some extent in explaining

participants’ purchasing habits being influenced by special occasions. The other two factors

identified albeit without greater emphasis were range/choice and store layout.

Price was an important factor for Sri Lankan participants but the degree of importance placed by

them was relatively low. For Sri Lankan participants, freshness and quality appeared to be

essential as some mentioned their unwillingness to compromise freshness and quality for price. As

a result of greater perceptions of freshness and quality, participants mentioned a preference to

shop at specialised stores for fresh produce; mirroring previous comments on specialist stores.

Further, they disclosed lower perceptions of freshness for such products at certain supermarkets

and a tendency to avoid buying fresh produce from such supermarkets as a result with one

participant mentioning;

120

I don’t buy most of the time not vegetables and fish from the supermarket because I don’t

believe, though they say it’s fresh, I don’t believe in that. (Male, aged 27: Sri Lanka)

Sri Lankan participants identified the importance of customer service. They were more concerned

about customer service afforded to them at supermarkets in contrast to UK participants who did

not appear to be as concerned about this element. Some participants mentioned their preference

towards a particular grocery retailer over another due to better standards of service and also

mentioned that poor customer service could result in them avoiding a retailer in the future;

I wouldn’t go again. Depending on the... what kind of thing happened, I wouldn’t

(Male, aged 24: Sri Lanka)

In contrast, none of the UK participants mentioned that they would require customer service while

some participants even mentioned that they would not be affected by poor service;

That doesn’t bother me. That wouldn’t stop me from going to a shop. If I need the stuff, I

need stuff... (Female, aged 26: UK)

This could be a result of a cultural difference in the form of Power Distance. Sri Lankan participants

who are high power distant may require greater customer service from grocery retail staff while

UK participants who are low power distant may perceive staff to be equal and as such, not expect

such customer service. Furthermore, UK participants viewed friendly and polite service cynically as

they believed such behaviour from staff was unnatural. On the contrary, Sri Lankan participants

welcomed such behaviour from staff. This again could be due to the difference in power distance

where Sri Lankan customers may expect staff to treat them with respect and courtesy. Further,

low Masculinity in the country may also influence such preferences in Sri Lanka; customers being

more prone to relationship building and open to such efforts. On the contrary, UK customers may

shun such behaviour from employees due to their Masculine cultural inclinations. Consequently,

customer service could have a greater effect in more power distant and feminine cultures such as

Sri Lanka than in less power distant and masculine cultures such as the UK.

Sri Lankan participants placed relatively more emphasis on range and choice. This could be a result

of the structure, where unlike in the UK, many stores in Sri Lanka are relatively small. As a result,

customers may look to shop at larger stores to ensure a better variety of products. Participants

also identified that Arpico and Keells provide a wider range of products; allowing them a “one-

stop-shop”.

121

We try to go to the big supermarkets as much as possible because there’s more variety and

choice because you end up going to a small outlet, something’s not available; you have to

go to another place (Female, aged 29: Sri Lanka)

I mostly drop in at Arpico or Keells for a “one-stop-buy”... (Male, aged 55: Sri Lanka)

These findings show that the constituents of Store Image may be valued differently in the two

countries. As identified, freshness and quality appear to be of greater importance to Sri Lankan

participants whilst price and convenience are highly valued by UK participants. The conceptual

framework (Figure 3.8) also identified the positive influence of service quality on customer loyalty.

Findings highlight how this would be more influential in Sri Lanka where greater expectations of

customer service appear to influence participants’ store preferences. The figure below summarises

these findings. Findings in red are exclusive to the UK and those in blue are exclusive to Sri Lanka.

Findings pertaining to both countries are in black. The (+) symbol denotes the greater influence of

customer service in Sri Lanka and the possible positive influence on customer loyalty. The

illustration below reflects the way it is depicted in the refined conceptual framework.

Figure 5.3: Key Attributes of Store Image Influencing Customer Loyalty

5.2.3 Differences in Customer Loyalty

UK participants mentioned a tendency for deal-proneness and cherry-picking as well as timing

their visits to pick up deals and offers. They were accustomed to such deals/offers and mentioned

how they would exploit the timing of such incentives. Participants went on to mention that they

would shop at specific stores for specific brands to make use of such deals and offers. UK

participants further mentioned that money-off vouchers issued by loyalty schemes influence their

shopping behaviour. Based on these comments, such offers appear to influence customer

Customer Loyalty

Store Image Price, Deals & Offers Loyalty programmes Quality Ethical considerations Convenience Store Layout Range & Choice Freshness Customer Service (+)

Key to Colour Coding Red- UK Blue- Sri Lanka Black- Both Countries

122

switching; resulting in divided patronage. Furthermore, holding several loyalty cards will enable

customers to receive more offers, resulting, yet again in them being prone to switching and be

more promiscuous in their purchasing behaviour. While such offers influence purchasing

behaviour, they may not enhance attitudinal loyalty. Similarly, this could make customers more

spurious as they may be more prone to switching as different loyalty schemes offer such deals. Sri

Lankan grocery retailers do not offer deals and offers to such an extent and it is probable that UK

customers as a result may be more spuriously loyal to their main grocery retailers. The possibility

of such behaviour was highlighted by the comment below;

Yeah I get sent um, offers because I have a Clubcard and um, Nectar as well so if we get

something through the post that says “eight pounds off sixty quid” on week A,B,C, then

uh... we use Sainsbury’s that week and then if there’s a... I get similar ones on the email

from Tesco’s. If there’s a good enough one for that week, then that usually sways me a

little bit (Male, aged 35: UK)

UK participants mentioned a greater favourable attitude towards Waitrose and Marks & Spencer

in relation to the other retailers. They expressed an aspiration to shop at these two retailers given

their favourable attitude; suggesting greater attitudinal loyalty towards these two retailers.

Despite this favourable attitude, several participants mentioned their inability to shop at the two

premium grocery retailers mainly due to higher prices with one participants saying;

That’s not where... I will shop purely out of choice... ‘cos you know constraints like time and

cost and things like that. If I had absolute choice, I’d probably shop at Waitrose

(Male, aged 26: UK)

As this favourable attitude is not backed by corresponding purchasing behaviour, such customers

could be latent loyals. However, they mentioned that they would shop at Marks & Spencer or

Waitrose if their financial situation allowed them to do so. If this favourable attitude is backed by

the ability to spend, such customers could develop true loyalty towards these grocery retailers.

This favourable attitude towards the two premium grocery retailers could be a result of their

superior reputation in terms of the quality of their products and the overall shopping experience.

Further, positioning strategies used by these retailers may have led to them being perceived as

premium grocery retailers. The greater quality perception attached to these two retailers was

highlighted by participants in one focus group discussion placing them well above the other

grocery retailers on a perceptual map (see Appendix H).

123

Despite these preferences, participants expressed their satisfaction with the quality of products

available at the other retailers; Big Four as well as Aldi and Lidl. Whilst acknowledging the superior

quality at Waitrose and Marks & Spencer, participants identified that the other main retailers also

uphold satisfactory standards of quality. This suggests that whilst holding preferences for the two

premium retailers, participants do in fact hold positive attitudes towards their main retailers. The

following statements confirm this and it can be argued that there is overall true loyalty to the main

grocery retailers in the UK whilst the two premium retailers enjoy latent loyalty.

Morrisons has a reputation for quality. I think that’s through their advertising, you know

they’re continuously promoting the freshness and the market style… (Female, aged 53: UK)

It’s quite a lot cheaper at Aldi for similar sort of quality for me personally anyway... (Female, aged 26: UK)

In stark contrast, Sri Lankan participants did not mention such aspirations; possibly due to the

nonexistence of high-end grocery retailers as well as premium positioning strategies. Some

participants however, mentioned a preference towards their usual grocery retailers and negative

perceptions towards others. Particularly those who regularly shopped at Keells and/or Arpico,

mentioned a preference towards those grocery retailers and a negative attitude towards Cargills

as well as Laugfs and Sathosa;

Otherwise planned stuff, never go to these two... (Male, aged 55: Sri Lanka)

Yeah, I will always uh, not go to (Cargills) Food City. Always try not to go to Food City...


These preferences and negative attitudes appeared to be a result of participants judging these

grocery retailers based on the quality and freshness of their products with special emphasis on

fresh produce as well as the level of customer service;

Don’t go to (Cargills) Food City or Laugfs if you want to buy meat or fish


Participants appeared to view Keells and Arpico as premium retailers compared to the rest of the

chains due to their perceptions of greater product quality and customer service. This is surprising

as these two retailers do not portray a premium image through their marketing communications.

124

As discussed previously, the importance placed on freshness and quality as well as customer

service is clearly evident. Consequently, Sri Lankan customers who shop at Keells and Arpico could

be deemed to be truly loyal due to their favourable attitudes as well as purchasing behaviour.

Negative attitudes towards Cargills as well as Laugfs and Sathosa suggest spurious loyalty towards

these retailers; despite repeat patronage, customers appear to lack attitudinal preferences

towards these retailers. It is also likely that these customers may have latent loyalty towards Keells

and Arpico. Such preferences may not be complemented by purchasing due to these two retailers

being more expensive. This was clearly identified by one participant;

I would love to go to Arpico but um… because of the price… (Female, aged 26: Sri Lanka)

It is clear that despite the lack of positioning strategies, these two retailers enjoy a similar position

with Waitrose and Marks & Spencer in the UK. Whilst positioning appears to drive preference

towards the premium retailers in the UK, retailer focus on customer service and product quality

appears to influence the same in Sri Lanka.

Overall, these findings suggest that whilst preferring premium retailers, UK participants in fact, are

truly loyalty to their main retailers. This is highlighted by the fact that they hold positive attitudes

towards the Big Four as well as two discounters. On the contrary, true loyalty is only apparent

towards Keells and Arpico in Sri Lanka whilst participants demonstrate spurious loyalty towards

Cargills and the other retailers. It can therefore be argued that greater overall true loyalty could be

expected in the UK whereas greater spurious loyalty as well as latent loyalty could be expected in

Sri Lanka.

Interestingly however, Sri Lankan participants mentioned a lack of better alternatives in grocery

retailing. In contrast to the UK, only five grocery retail chains operate in the country. As a result,

customers may feel deprived of a full range of grocery retailers. Further, given their negative

perceptions of some retail chains, they may feel even more starved of alternatives. This could

result in them being confined to a particular retailer. This was mentioned by participants whose

main supermarket was either Keells or Arpico with one participant saying; “There is no one else

better than this...” (Male, aged 26: Sri Lanka). Conversely, UK participants mentioned that the

availability of a greater choice of grocery retailers allows them to switch more freely. Given the

lack of alternatives, more captive loyalty as conceptualised by Rowley (2005) could be expected in

125

Sri Lanka. More promiscuity could be prevalent in the UK due to greater choice which may lead to

switching behaviour.

These findings clearly highlight differences in loyalty between the two countries. As illustrated by

Figure 5.2, greater promiscuity may be prevalent in the UK due to customers focusing on price,

deals and offers. However, participants’ favourable attitudes towards their main store suggest

overall true loyalty in the UK. Greater spurious loyalty could be expected in Sri Lanka due to the

lack of attitudinal loyalty to the market leader; Cargills as well as Laugfs and Sathosa. However,

preferences for the two premium retailers in the UK as well as Keells and Arpico in Sri Lanka

suggest that there may be a degree of latent loyalty in the two countries; with this being greater in

Sri Lanka. As summarised in Figure 5.4, findings in red are exclusive to the UK and those in blue are

exclusive to Sri Lanka. The (+) and (-) symbols denote the possible positive or negative implications

for customer loyalty. The illustration below reflects the way it is depicted in the refined conceptual

framework.

Figure 5.4: Key Findings pertaining to Differences in Customer Loyalty

Customer Loyalty

Key to Colour Coding Red- UK Blue- Sri Lanka

Latent Loyalty (-) Spurious Loyalty (-)

True Loyalty (+) Captive Loyalty (-)

126

5.2.4 Loyalty Programmes

As the discussion progressed towards the role of loyalty programmes, UK participants shared their

concerns regarding the tracking of customer behaviour. They also mentioned feeling

uncomfortable as a result of being tracked and some of them mentioned feeling manipulated by

grocery retailers who use their personal and behavioural data to target them;

But it makes me feel a little bit uncomfortable and I’m aware of it, I’m aware I’m being

manipulated to buy... (Female, aged 37: UK)

Feeling uncomfortable may be a result of them considering such actions to be an intrusion of their

privacy. They also shared concerns about the way their data are used by retailers and mentioned

that they expect retailers to be more transparent in this regard. This was highlighted by one

participant who suggested that small independent grocers were more trustworthy in this regard;

They wouldn’t really be selling my data over... (Male, aged 31: UK)

In contrast, Sri Lankan participants mentioned the lack of a proper loyalty scheme and a lack of

awareness. They also highlighted the need for grocery retailers to track their behaviour so that

more customised offers could be provided. Given the lack of such practices in Sri Lanka, these

participants expressed their openness to being tracked. Participants also identified that although

grocery retailers track their behaviour, nothing was being done using their data. These thoughts

are typified by the following comments;

I think... they should you know track us down and you know do a research on our patterns

and then you know come up with offers… (Male, aged 27: Sri Lanka)

I know that they track us already but it’s like just tracking us... give us the benefit of that

tracking. We don’t see that... (Male, aged 31: Sri Lanka)

Furthermore, given the cultural difference in the form of Individualism and Collectivism,

collectivist Sri Lankan customers may be more open to such initiatives than their UK counterparts.

This could be due to individualistic customers valuing personal space which may lead to them

being averse to tracking. Furthermore, the feminine cultural inclinations in Sri Lanka may make

customers more prone to relationship building and in turn, be open to such efforts from

businesses. A structural aspect is also evident. UK customers are aware of tracking due to the

widespread nature of such initiatives. They may be averse to tracking as they are aware of how

127

businesses use their data; resulting in pessimism. On the other hand, Sri Lankan customers may be

open to such tracking as they are not as experienced with their data being gathered and used by

businesses. A very good example is the fact that UK participants rejected the idea of being greeted

or receiving a gift on a special occasion such as a birthday or an anniversary whereas the Sri

Lankan participants welcomed this idea. In fact, this was mentioned by one participant without

being prompted while others mentioned being open to such initiatives;

If they’re very loyal; track the birthday, anniversaries, yeah... (Male, aged 55: Sri Lanka)

The aversion to such initiatives in the UK was typified by the following comment;

You getting a cake or a card from Tesco’s is like... that is what one person says to another,

it’s not a big corporation (Male, aged 26: UK)

In relation to loyalty scheme benefits, both countries’ participants mentioned their expectation of

a wider range of redemption options instead of simply redeeming their loyalty points in-store to

buy groceries. Sri Lankan participants appeared to be more concerned about redemption options

due to the lack of adequate redemption options as identified by them. UK participants mentioned

their appreciation of Tesco’s Clubcard in particular because of the wider range of redemption

options available. For Sri Lankan participants, benefits were basically in the form of discounts on

grocery products which they were not too keen on; “they don’t give us money for those points,

they give us groceries...” (Male, aged 34: Sri Lanka). When probed as to whether participants

would be attracted to grocery retailers providing more redemption options, one participant said;

“something like that would be good...” (Male, aged 28: Sri Lanka) with others agreeing;

Then we would definitely go there and would also definitely use the card...


Furthermore, both countries’ participants mentioned a preference for instant or segregated

rewards over delayed or aggregated rewards. In line with this preference, participants in both

countries expressed a lack of willingness to accumulate points for future rewards;

We may not accumulate points but it may stay in the wallet... (Male, aged 28: Sri Lanka)

This could be explained by the low Long Term Orientation culture in both countries where they

would seek more short term gains. However, Sri Lankan participants appeared to be more open to

128

soft benefits than UK participants. This can be supported by their openness to approaches from

grocery retailers on special occasions as discussed previously. Their high Collectivism as well as

their feminine culture may make them more prone to such initiatives and relationship building

efforts compared to UK customers.

Another interesting finding is that Sri Lankan participants expect preferential treatment as

members of a loyalty programme which non-members would not receive. These participants also

mentioned a liking towards a tiered loyalty programme where higher tiers would enable them to

obtain greater benefits as mentioned by one participant;

But if the Privilege card or the loyalty card scheme would get me ahead of in the line, I

would... I’ll shop more in that particular shop to build up the points if that be the case...


The need for such treatment and tiered schemes could again be explained by the high Power

Distance culture which focuses on status. UK participants did not mention such preferences;

suggesting a lack of openness to preferential treatment and tiered schemes. The status seeking

nature of Sri Lankan customers is aptly illustrated by the following comments and highlights that

such expectations permeate throughout other retail contexts in the country;

You have a VIP line where you can just... If they had levels of cards, you know you have the

gold Privilege card at Arpico, you come here, you pay, you go; that’s it...


In Amex… it starts with the gold… Because people have the perception where you have a

gold card, OK, I’m like, I’m a king so… just to give that, uh… satisfaction


Clear differences between the two countries in respect of participants’ expectations of loyalty

programmes are evident with one similarity. Whilst participants in both countries would prefer

instant or segregated rewards, Sri Lankan participants would prefer soft benefits, preferential

treatment and tiered schemes as well as being open to being tracked. The figure below

summarises these findings; findings in red are exclusive to the UK and those in blue are exclusive

to Sri Lanka. Findings pertaining to both countries are in black whilst the (+) and (-) symbols

denote the possible positive or negative implications for customer loyalty.

129

Figure 5.5: Key Findings pertaining to Loyalty Programmes

5.2.5 Section Summary & Refined Conceptual Framework Development

The refined conceptual framework illustrated by Figure 5.6 summarises the key qualitative findings

presented in this section. Findings in red are exclusive to the UK and those in blue are exclusive to

Sri Lanka. Findings pertaining to both countries are in black whilst the (+) and (-) symbols denote

the possible positive or negative implications for customer loyalty.

Section 5.2.1 discussed the findings pertaining to grocery shopping habits and the key findings are

summarised in Figure 5.2. Special occasions as well as deal proneness and promiscuity appear to

have a negative influence on behavioural loyalty in the UK whilst preference for specialist stores in

both countries may result in a similar influence on behavioural loyalty. Aversion to own labels in

Sri Lanka and negative attitudes towards retailers promoting such products over national brands

may cause negative attitudinal loyalty to such retailers. Key findings pertaining to Store Image are

discussed in section 5.2.2 and it was identified that customer service may be more important in Sri

Lanka whilst the other constituents would be valued differently in the two countries. These are

summarised in Figure 5.3.

Section 5.2.3 identified the differences in customer loyalty and it was identified that greater true

loyalty may be prevalent in the UK whilst greater spurious loyalty and captive loyalty could be

prevalent in Sri Lanka. Latent loyalty to premium retailers was identified in both countries but this

appeared to be greater in Sri Lanka. These findings are summarised in Figure 5.4. Finally,

differences in perceptions of loyalty programmes are discussed in section 5.2.4 and the key

findings are summarised in Figure 5.5.

Loyalty Programmes

Soft Benefits (+)

Instant Rewards (+)

Tiered Schemes (+)

Preferential Treatment (+)

Tracking (-) Tracking (+)

Customer Loyalty


130

The refined conceptual framework presented below summarises the key findings discussed in sections 5.2.1 to 5.2.4. This framework is a combination

of Figures 5.2 to 5.5 (summarising sections 5.2.1 to 5.2.4).

Figure 5.6: Refined Conceptual Framework

Loyalty Programmes (Section 5.2.4; Figure 5.5) Special Occasions (-)

Deal Proneness(-)

Specialist Stores (-)

Customer Loyalty

Own Labels (-)

Grocery Shopping Habits

(Section 5.2.1; Figure 5.2)

Soft Benefits (+)

Instant Rewards (+)

Tiered Schemes (+)

Preferential Treatment (+)

Tracking (-) Tracking (+)

Store Image (Section 5.2.2; Figure 5.3) Price, Deals & Offers Loyalty programmes Quality Ethical considerations Convenience Store Layout Range Freshness Choice Customer Service (+)

Differences in Customer Loyalty (Section 5.2.3; Figure 5.4)

Latent Loyalty (-) Spurious Loyalty (-)

True Loyalty (+) Captive Loyalty (-)


131

Section 2

Quantitative Results

132

5.3 Second Empirical Study- Questionnaires

This section presents the results of the various analyses carried out to further investigate the

refined conceptual framework. Characteristics of the two countries’ samples will be described

first, followed by the results of the various analyses.

5.3.1 Sample Characteristics- UK

The characteristics of the sample are summarised in Table 5.2.

Description Number Percentage of Sample Gender Female Male Missing

205 68 1

74.8% 24.8% 0.4%

Age 18-24 25-34 35-44 45-54 55-64 65-74 75 and above Missing

9

28 28 63 48 57 29 12

3.3%

10.2% 10.2% 23%

17.5% 20.8% 10.6% 4.4%

Ethnic Origin White Mixed Asian or Asian British Black or Black British Chinese Missing

259

1 3 6 2 3

94.5% 0.4% 1.1% 2.2% 0.7% 1.1%

Annual Household Income Below £15,000 £15,000 – £19,999 £20,000 – £29,999 £30,000 – £39,999 £40,000 – £49,999 Greater than £50,000 Missing

62 27 33 38 26 67 21

22.6% 9.9% 12%

13.9% 9.5%

24.5% 7.7%

Highest Academic Qualification Up to GCSE GCSE or Equivalent A Level or Equivalent First Degree Master’s Degree or Higher Other Missing

31 34 59 76 38 22 14

11.3% 12.4% 21.5% 27.7% 13.9%

8% 5.1%

Table 5.2: Characteristics of the UK sample (n = 274)

Whilst the gender split between females and males in the UK is 50.7% to 49.3% (Office for

National Statistics 2017b), an overrepresentation of females is evident. This is possibly because

women are more likely to be responsible for all or most of their household’s food and grocery

shopping; 68% compared to 32% of men (Food Standards Agency 2016). While the sample

133

adequately represents each adult age group, it is evident that the age groups of 35-44, 45-54

and 55-64 are overrepresented in the sample compared to the UK population (Office for

National Statistics 2017b). While the age groups of 18-24 and 65-74 are somewhat evenly

represented, the age groups of 25-34 and 75+ are underrepresented.

The household income distribution within the sample is similar to that of the population

(Office for National Statistics 2018) with the average annual household income in the £35,000

region. The sample represents the low qualification groups in the UK somewhat equally with

45.2% representation compared to 55% in the population (Office for National Statistics 2012).

However, 41.6% of the sample represent high qualification groups (first degree and above)

compared to 27% in the population. 94.5% of the respondents in the sample are White British

compared to 87.2% in the UK (Institute of Race Relations 2018) while the other ethnic groups

are underrepresented.

5.3.2 Sample Characteristics- Sri Lanka

The gender split between females and males in Sri Lanka is 51.6% to 48.4% (Department of

Census and Statistics- Sri Lanka 2018a) and the split in the sample is not too dissimilar. Whilst

the adult age groups are adequately represented in the sample, the 25-34 age group is

overrepresented compared to the Sri Lankan population (Department of Census and Statistics-

Sri Lanka 2018b) with this age group constituting 47.1% of the sample. While the age groups of

18-24, 35-44, 45-54 and 55-64 are somewhat evenly represented in the sample, the 65-74 age

group is underrepresented compared to the population, while none of the respondents are

aged above 75, who; due to multi-generational living in Sri Lanka were not of interest to this

study since they rarely conduct the main grocery shopping.

Although the household income distribution within the sample is dissimilar to that of the

population (Department of Census and Statistics- Sri Lanka 2017), the different income groups

are adequately represented in the sample. Although academic qualification data is not

available for Sri Lanka, 53.6% of the sample was educated above first-degree level. While the

representation of Sinhalese and Tamil respondents in the sample is somewhat similar to that

of the population (Department of Census and Statistics- Sri Lanka 2018a), Muslims are

underrepresented.

The characteristics of the sample are summarised in Table 5.3.

134

Description Number Percentage of Sample Gender Female Male Missing

164 110

2

59.4% 39.9% 0.7%

Age 18-24 25-34 35-44 45-54 55-64 65-74 75 and above Missing

18

130 45 37 32 5 0 9

6.5%

47.1% 16.3% 13.4% 11.6% 1.8% 0%

3.3% Ethnic Origin Sinhalese Tamil Muslim Burgher Other Missing

209 46 9 8 2 2

75.7% 16.7% 3.3% 2.9% 0.7% 0.7%

Annual Household Income Below £2,039 £2,039 – £4,077 £4,077 – £6,115 £6,116 – £8,154 £8,154 – £10,192 Greater than £10,193 Missing

23 64 54 40 23 62 10

8.3%

23.2% 19.6% 14.5% 8.3%

22.5% 3.6%

Highest Academic Qualification Up to GCSE GCSE or Equivalent A/L or Equivalent First Degree Master’s Degree or Higher Other Missing

9

13 88 96 52 12 6

3.3% 4.7%

31.9% 34.8% 18.8% 4.3% 2.2%

Table 5.3: Characteristics of the Sri Lankan sample (n = 276)

5.3.3 Preliminary Composition of Items: Exploratory Factor Analysis

The preliminary composition of the items used to measure the various constructs was

ascertained through separate exploratory factor analyses in the two countries. Satisfactory

KMOs and Bartlett’s Test of Sphericity confirmed the suitability of the data for these analyses.

The composition of the factors extracted in both countries confirms that the items used to

measure the various constructs represent the respective constructs accurately as intended. As

a result, the pattern matrices can be used for confirmatory factor analysis with greater

confidence.

10 factors and 11 factors were extracted in the UK and Sri Lanka respectively. REW (Rewards)

was the 11th factor extracted which was not included in the UK pattern matrix. The construct;

135

Special Occasions was not extracted in either country due to poor factor loadings. Overall

Alpha scores for all the factors in both countries were above the threshold of 0.7 (Nunnally

1978). No Corrected Inter-Item Correlations below 0.3 were observed; thus, negating the need

to remove any variables. The convergent and divergent validity of the factors was also

ascertained. The factors were then named in correspondence with the refined conceptual

framework (Figure 5.6). Appendices I and J provide details of these factors in terms of their

eigenvalues, percentages of variance explained and Cronbach’s Alpha scores for the United

Kingdom and Sri Lanka respectively. The final pattern matrices were then used to run separate

confirmatory factor analyses for the two countries. The names assigned to each factor are as

follows;

Factor 1 True Loyalty Factor 2 Openness to Preferential Treatment Factor 3 Spurious and Latent Loyalty Factor 4 Openness to Soft Benefits Factor 5 Aversion to Supermarket Own Labels Factor 6 Importance of Customer Service Factor 7 Openness to Tiered Schemes Factor 8 Captive Loyalty Factor 9 Aversion to Fresh Produce in Supermarkets Factor 10 Aversion to Tracking 5.3.4 Confirmatory Factor Analysis

Following on from the exploratory factor analyses, the reliability and validity of the constructs

confirm that the items used in fact, accurately measure the constructs as intended. This

confirms the robustness of the questionnaire items as well as the data. This is important as all

the items except for those measuring True Loyalty were exploratory or qualitative in nature. In

other words, these items were not borrowed from previous studies; but developed based on

the qualitative findings.

Table 5.4 presents the final model fit figures for the UK and Sri Lanka. The composite reliability,

convergent validity and divergent validity figures for the United Kingdom and Sri Lanka are

presented in appendices K and L respectively. The model fit figures for both countries are

satisfactory based on the thresholds proposed by Hair et al. (2010) and Gaskin (2017c).

Measure UK Sri Lanka Threshold Chi-square/df (cmin/df) 1.694 1.712 < 5 or ideally < 3 p-value of the model <.000 <.000 > .05 (Not applicable with large samples) CFI .941 .930 > .90 or ideally > .95 SRMR .051 .056 < .09 RMSEA .050 .051 Between 0.03 and 0.08 PCLOSE .433 .375 > .05 Table 5.4: Model Fit for UK and Sri Lanka

136

The measurement model in AMOS version 22 with all the variables in the UK pattern matrix

yielded overall satisfactory reliability and validity figures with two constructs; Aversion to Fresh

Produce in Supermarkets and Aversion to Supermarket Own Labels having convergent validity

figures well below the 0.4 threshold (Bagozzi and Yi 1988). The initial measurement model with

the Sri Lankan pattern matrix did not yield satisfactory convergent validity figures for five

factors; Aversion to Tracking, Aversion to Supermarket Own Labels, Importance of Customer

Service, Captive Loyalty and Aversion to Fresh Produce in Supermarkets. Aversion to

Supermarket Own Labels and Captive Loyalty had the lowest average variance extracted thus

two variables; ‘I am sceptical about the quality of own label products’ and ‘I would switch if a

suitable alternative supermarket(s) became available’ were removed. This resulted in the

improvement of the convergent validity of these factors.

The reliability and validity of the constructs in the UK are satisfactory. Composite reliability was

greater than 0.6 and average variance extracted was greater than 0.5 (Bagozzi and Yi 1988). It

should be noted however, that this threshold is too stringent (Malhotra and Dash 2011). The

square root of the average variance extracted is greater than the corresponding inter-

construct correlations (Fornell and Larcker 1981; Anderson and Gerbing 1988). The composite

reliability and divergent validity of the constructs in Sri Lanka are also satisfactory. The

construct scores were then imputed.

5.3.5 Inter-Country Differences

This section presents the results of two analyses; the comparison of the two countries’

constructs (imputed construct scores post-CFA) and the comparison of the importance placed

on attributes of store image between the two countries.

5.3.5.1 Comparison of Constructs

Table 5.5; highlights the key differences between the two countries as identified through the

Mann-Whitney U Tests. Significant differences (p < .05) can be observed between 8 of the 10

constructs. The difference between the two countries in respect of Aversion to Fresh Produce

in Supermarkets (p= .827) and Importance of Customer Service (p= .547) is not significant.

Whilst there is a low level of aversion to fresh produce in both countries, customer service is

highly important as reported by the two countries’ respondents. Follow-up median scores and

effect sizes were calculated for the 8 constructs with significant differences.

137

Construct UK Median SL Median Sig. Effect Size (r) Grocery Shopping Habits (Section 5.2.1)

Aversion to Supermarket Own Labels 3.266 1.422 <.001 0.76 Loyalty Programme Attributes (Section 5.2.3)

Openness to Soft Benefits 2.249 3.806 <.001 0.62 Openness to Tiered Schemes 1.622 3.439 <.001 0.62 Aversion to Tracking 4.102 2.432 <.001 0.53 Openness to Preferential Treatment 2.569 3.827 <.001 0.43

Customer Loyalty (Section 5.2.3) Spurious and Latent Loyalty 2.411 3.044 <.001 0.18 True Loyalty 4.419 4.179 .001 0.14 Captive Loyalty 1.734 2.161 .018 0.10 Table 5.5: Differences between the UK and Sri Lanka (UK = 274, SL = 276)

A significantly greater aversion towards supermarket own labels amongst UK respondents is

evident with a large effect further substantiating this difference. It is also interesting to note

that while the overall median in the UK is above 3, a significantly low overall median is

reported for Sri Lanka which suggests that Sri Lankan customers are in fact not averse to

supermarket own label brands. This finding stands in stark contrast to the qualitative findings

where a high degree of aversion was expressed by Sri Lankan participants whilst such

perceptions were not evident from UK participants. This result albeit contradictory, provides

an interesting insight in both countries. Despite the pervasiveness of such products in the UK,

customers are in fact moderately averse to such products. The reported openness to own

labels in Sri Lanka is again intriguing, given the relative novelty of such products.

UK respondents are significantly more averse to being tracked by loyalty schemes compared to

their Sri Lankan counterparts. Sri Lankan respondents are significantly more open to tiered

schemes, soft benefits and preferential treatment. This greater openness is substantiated by

medium to large effect sizes. On the contrary, low scores in the UK suggest an aversion to such

initiatives. Overall, Sri Lankan respondents are significantly more open to the four loyalty

programme attributes.

Although significantly greater in Sri Lanka, a low degree of captive loyalty is reported in both

countries. The overall medians for the two countries in this respect are low and the effect size

further denotes little difference between the two countries. The result shows that regardless

of structural differences in terms of the number of stores, customers in both countries are

content with the availability of grocery retailers. The degree of true loyalty is significantly

greater in the UK albeit with a small effect. Significantly greater spurious and latent loyalty is

also reported in Sri Lanka. Despite the statistically significant difference in true loyalty between

the two countries, the effect is low and should be noted that high levels of true loyalty can be

observed in both countries. The degree of spurious and latent loyalty in Sri Lanka is somewhat

138

moderate as suggested by the overall medians whereas a lesser degree can be observed in the

UK given its lower overall median.

5.3.5.2 Store Image

Table 5.6 shows the differences in the degree of importance placed on each attribute of Store

Image between the two countries. The Mann-Whitney U Test yielded significant differences for

20 of the 24 attributes. r values in bold represent medium and large effect sizes. The results

show that customers in both countries generally place similar degrees of importance on

attributes of Store Image. This is confirmed by the difference in medians for 17 attributes

being 0 or 1 with 13 of the corresponding r values denoting small effects. However, Affiliated

Credit Cards, One-Stop-Shop and Payment Options are of much greater importance to Sri

Lankan respondents as highlighted by the large or medium effect sizes. Of particular interest is

the higher overall median reported for ethical considerations in Sri Lanka followed by a

medium effect; highlighting the importance of these considerations to customers. This is

interesting as ethical considerations were not identified by participants in the Sri Lankan focus

groups. It should also be noted that whilst freshness and quality appeared to be of greater

importance to Sri Lankan participants, respondents in both countries place a high level of

importance on these attributes.

Attribute Asymp. Sig. (2-tailed)

UK Median

SL Median Effect Size (r)

Affiliated credit cards <.001 2 6 0.61 One-stop-shop <.001 6 7 0.46 LP offers <.001 4 6 0.41 LPs <.001 4 5 0.36 Ethical considerations <.001 5 6 0.35 Convenience <.001 6 7 0.34 Payment options <.001 5 7 0.34 Customer service <.001 6 7 0.28 Accessibility <.001 6 7 0.23 Opening times <.001 6 6 0.20 Trust <.001 6 7 0.20 Store layout <.001 5 6 0.19 Range & choice <.001 6 7 0.18 Distance to home/work <.001 6 6 0.16 Deals & offers 0.002 5 6 0.14 Parking 0.002 6 7 0.14 Reputation 0.003 6 6 0.13 VFM 0.011 6 7 0.11 Freshness 0.046 7 7 0.09 Quality 0.048 7 7 0.09 Table 5.6: Degree of importance placed on attributes of Store Image (UK = 266, SL = 237)

139

5.3.5.3 Store Usage & Customer Promiscuity

59% of Sri Lankan respondents reported using only 1 or 2 supermarkets per month compared

to 28.5% in the UK. Contrastingly, more UK respondents reported using 3 or more

supermarkets per month; 70.1% of the sample compared to only 36.6% in Sri Lanka. Only

13.4% of Sri Lankan respondents reported using 4 or more supermarkets monthly compared to

38.7% of UK respondents. These descriptive statistics show that UK customers use more stores

per month compared to their Sri Lankan counterparts. Hence, greater promiscuity, by way of

store usage is evident in the UK. This was confirmed by a follow-up Chi-square test for

independence (with Yates Continuity Correlation). The Chi-square test after categorisation

(Low- 1 or 2 supermarkets and High- 3 or more supermarkets) indicated a significant difference

between the two countries in this respect; X2 (1, n= 534), p <.001, phi = -.285. The phi

coefficient indicates a medium effect (Pallant 2016).

5.3.6 Within-Country Differences: United Kingdom 5.3.6.1 Differences in Customer Loyalty towards Main Supermarket in the UK

Significant differences in true loyalty (p= .001) as well as spurious & latent loyalty (p= .003)

exist between the four supermarket categories. Captive loyalty does not vary (p= .160) across

these categories. Table 5.7 shows the results of the post-hoc Mann-Whitney U Tests with

Bonferonni corrections as well as follow-up median scores and effect sizes for paired tests that

are significantly different (p <.0083). This significance level for acceptance was set given the

use of 4 categories, requiring 6 paired tests (0.05/6).

These results confirm that the two Premium retailers enjoy the greatest true loyalty from

customers while the two Discounters are most affected by spurious & latent loyalty. However,

a high level of true loyalty can be observed (overall median > 4) towards all four supermarket

categories; confirming that UK respondents in general, are truly loyal to their main retailer.

Similarly, the degree of spurious & latent loyalty towards these four categories, Big Four and

Discounters in particular is not too severe (overall median < 3). The results also confirm that

the two premium retailers attract the most latent loyalty from customers. This is highlighted

by the significantly greater levels of spurious loyalty reported towards the Big Four retailers

and the two Discounters.

140

Description Asymp. Sig. (2-tailed)

Median 1 Median 2 Effect Size (r)

Big Four versus Discounters True Loyalty Spurious & Latent Loyalty

0.023 0.781

Median scores and effect sizes are not applicable as the differences are insignificant

Big Four versus Premium True Loyalty Spurious & Latent Loyalty

<.001 <.001

4.295 2.439

5.184 1.335

0.26 0.26

Big Four versus Other True Loyalty Spurious & Latent Loyalty

.276 .735


Discounters versus Premium True Loyalty Spurious & Latent Loyalty

.021 .002

N/A

2.629

N/A

1.335

N/A 0.35

Discounters versus Other True Loyalty Spurious & Latent Loyalty

.979 .857


Premium versus Other True Loyalty Spurious & Latent Loyalty

.311 .022


Key: Big Four: Tesco, Sainsbury’s, Asda & Morrisons Discounters: Aldi & Lidl Premium: Marks & Spencer and Waitrose Other: The Co-op, Iceland and other Table 5.7: Differences in customer loyalty towards the main supermarket (N = 274)

5.3.6.2 Supermarket Availability and Customer Loyalty in the UK

The result indicates that low availability conditions result in greater captive loyalty amongst

supermarket customers. The results of the Mann-Whitney U Test involving Captive Loyalty and

the level of availability (Low- 0, 1 and 2 versus High- 3, 4 and 5 or more) show a significant

difference between the two availability categories (p < .05). Table 5.8 provides the test

statistics, follow-up median scores and effect size. A significant difference can be observed

between the availability categories with a small effect reported. However, low overall medians

are reported for both categories; reflecting the low overall captive loyalty in the UK (Overall

Md = 1.736).

Asymp. Sig. (2-tailed)

Median Low Availability

Median High Availability

Effect Size (r)

Captive Loyalty .002 1.903 1.666 0.19 Table 5.8: Supermarket availability and customer loyalty (N = 270)

141

5.3.6.3 Supermarket Usage and Customer Loyalty in the UK

The results highlight that there is greater true loyalty amongst respondents using fewer (1 or

2) supermarkets monthly. The results of the Mann-Whitney U Test focusing on the degree of

loyalty in relation to the monthly usage of supermarkets (Low- 1 and 2 versus High- 3, 4 and 5

or more) show a significant difference between the two usage categories in respect of true

loyalty. Table 5.9 provides the test statistics, follow-up median scores and effect size. Despite

the significant difference in respect of true loyalty, it should be noted that only a slight

difference exists in terms of the overall medians of the two usage categories as a high degree

of true loyalty is prevalent across both usage categories. This result highlights that despite

greater promiscuity being identified previously, customers are in fact truly loyal to their main

store.


Median Low Usage

Median High Usage

Effect Size (r)

True Loyalty .015 4.603 4.336 0.15 Table 5.9: Supermarket usage and customer loyalty (N = 270)

5.3.6.4 Content Analysis of Open-Ended Questions

Price and convenience being cited the most shows that these two factors are the most valued

by grocery shoppers. The overall dominance of price and other related aspects (discounts,

special offers and value for money) is evident by 67% reporting at least one element. Similarly,

the focus on convenience related aspects (convenience, proximity, parking, accessibility and

opening times) is evident by 79% reporting at least one element. Despite the pervasiveness of

loyalty programmes in UK grocery retailing, only 3% reported this element. Despite customer

service possessing an overall median of 6 in respect of store image attributes, only 1%

reported this attribute. 34% reported choice, variety and range as being an important factor;

further supporting the previous results. The most important reasons for the monthly usage of

supermarkets are summarised in Table 5.10.

Reason Number of Respondents Percentage of Respondents Price 98 47% Convenience 88 42% Choice, Variety & Range 72 34% Proximity 62 30% Quality 34 16% Special Offers 27 13% Availability 18 9% Value for Money 14 7% Table 5.10: Reasons for Monthly Supermarket Usage

142

The other content analysis focused on the open-ended responses provided to the two

questions at the end of Q10; name of supermarket(s) preferred over main supermarket and

reason(s). The 274 cases were sorted according to spurious & latent loyalty and cases with a

construct score greater than 2.4 (2.4 being the median for this construct). 138 cases exceeded

this threshold but 63 were removed due to missing data. 38 of the 75 eligible cases (51%)

reported Waitrose or Marks & Spencer as their preferred grocery retailer. 24 of the 51

respondents (47%) who reported a Big Four supermarket as their main supermarket

mentioned Waitrose or Marks & Spencer or both as their preferred alternative. Similarly, 11 of

the 19 respondents (58%) shopping at Aldi or Lidl reported either or both of the two premium

supermarkets as their preferred choice. This further supports previous findings that the two

Premium retailers are most preferred by UK respondents and enjoy the greatest degree of

latent loyalty.

Respondents reporting Waitrose or Marks & Spencer as their preferred choice alluded to the

perceived better quality, range of produce and customer service; “Better quality, more variety,

home delivery and customer service” (Female: 45-54). Some respondents also mentioned the

greater ethical standing of these two grocery retailers; “Ethical policy, sourcing, organics and

welfare” (Female: 45-54). However, the inability to shop at these high-end supermarkets due

to financial constraints was highlighted by several respondents; “I like their food but can’t

afford it” (Female: 35-44). Furthermore, while mentioning that these supermarkets are more

expensive; “Nice food but too expensive” (Female: 18-24), some respondents reported not

receiving sufficient value for money; “not worth the extra cost” (Female: 25-34) and “Just too

expensive, no value for money” (Female: 45-54). Perceptions of inadequate value for money

suggest a gulf between the high prices charged by these supermarkets and the perceived value

attached to their offerings.

Some respondents preferring the “Big Four” supermarkets also mentioned the high quality

perception attached to the high-end supermarkets but further stated that they could obtain

better value for money by shopping at a “Big Four” supermarket; “I love Waitrose’s quality but

I love Morrison's prices” (Male: 65-74). Respondents preferring Aldi or Lidl referred to aspects

such as being inexpensive while offering good quality; “Aldi is good for quality food at half the

price using their own brands” (Female: 25-34). However, these value driven supermarkets not

having as full a range as the “Big Four” supermarkets was reported by some respondents; “I

accept I can’t buy everything I want at Aldi” (Female: 55-64).

143

5.3.7 Within-Country Differences: Sri Lanka 5.3.7.1 Differences in Customer Loyalty towards Main Supermarket in Sri Lanka

Significant differences in true loyalty (p <.001) as well as spurious & latent loyalty (p <.001)

exist between the three supermarkets categories. Captive loyalty does not vary (p= .052)

across these categories. These results confirm that Keells and Arpico enjoy the greatest degree

of true loyalty from customers while Laugfs, Sathosa and other supermarkets are most

affected by spurious & latent loyalty. The moderate levels of true loyalty and spurious/latent

loyalty reported for Cargills as well as Laugfs and Sathosa confirm that these retailers have a

largely spurious customer base. It is also clear that customers of these retailers have latent

loyalty towards Keells and Arpico. Although a satisfactory level of true loyalty can be observed

(overall median > 3.6) towards all three retailer categories, it is evident that this is largely

influenced by the high level of true loyalty enjoyed by Keells and Arpico.

Table 5.11 shows the results of the post-hoc Mann-Whitney U Tests with Bonferonni

corrections as well as follow-up median scores and effect sizes for paired tests that are

significantly different (p < .017). This significance level for acceptance was set considering the 3

categories, requiring 3 paired tests (0.05/3).

Description Asymp. Sig. (2-tailed)

Median 1

Median 2

Effect Size (r)

Cargills versus Better True Loyalty Spurious & Latent Loyalty

<.001 <.001

3.921 3.191

4.367 2.395

0.32 0.23

Cargills versus Other True Loyalty Spurious & Latent Loyalty

.283

.209


Better versus Other True Loyalty Spurious & Latent Loyalty

.004 .017

4.367 2.395

3.651 3.608

0.24 0.20

Key: Cargills (Market Leader) Better: Keells and Arpico Other: Laugfs, Sathosa and other Table 5.11: Differences in customer loyalty towards the main supermarket (N = 258)

5.3.7.2 Supermarket Availability and Customer Loyalty in Sri Lanka

The results indicate that captive loyalty is not affected by the availability conditions (Low- 0, 1

and 2 versus High- 3, 4 and 5 or more) in the country (p= .213). However, high availability

conditions result in greater true loyalty as well as lower levels of spurious & latent loyalty.

Table 5.12 provides the test statistics, follow-up median scores and effect sizes.

144


Median Low Availability

Median High Availability

Effect Size (r)

True Loyalty .001 3.879 4.302 0.20 Spurious & Latent Loyalty .027 3.187 2.696 0.14 Table 5.12: Supermarket availability and customer loyalty (N = 268)

The degree of true loyalty is significantly greater in high availability conditions albeit with a

small effect. It is probable that attitudinal loyalty would increase as the availability of

supermarkets increases. This may be because customers will have an optimum choice of

grocery retailers where they are able to shop at their preferred retailer although this cannot be

statistically tested. This is supported by the significant difference between the two availability

categories in respect of spurious & latent loyalty where greater spurious & latent loyalty is

evident in low availability conditions. Spurious & latent loyalty, characterised by a lack of

attitudinal loyalty towards the main focal retailer suggests that low levels of attitudinal loyalty

in low availability situations could be attributed to limited choice.

5.3.7.3 Supermarket Usage and Customer Loyalty in Sri Lanka

The results show that true loyalty, spurious & latent loyalty or captive loyalty is not

significantly different in low and high usage conditions.

5.3.7.4 Content Analysis of Open-Ended Questions

The most widely cited reasons for the monthly usage of supermarkets are summarised in Table

5.13.

Reason Number of Respondents Percentage of Respondents Convenience 87 37% Freshness & Quality 70 30% Choice, Variety & Range 59 25% One-Stop-Shop 43 18% Availability 41 17% Price 39 17% Accessibility 34 14% Parking 33 14% Proximity 31 13% Customer Service 31 13% Table 5.13: Reasons for Monthly Supermarket Usage

89% of the respondents mentioned at least one element (convenience, accessibility, parking

proximity, time taken and opening times) related to convenience which shows the magnitude

of importance placed on this aspect. Compared to 67% of the UK respondents reporting price

related aspects, only 32% mentioned at least one related element (price, discounts, special

offers and credit card offers) in Sri Lanka. 18% reported One-Stop-Shop or being able to buy

everything under one roof as being important. Similarly, 17% reporting availability as a factor

145

further supports the need for greater availability, preferably within one supermarket store. In

comparison to only 1% reporting customer service as a factor in the UK, 13% reported this

attribute. Overall, these results complement the previous quantitative findings.

For the analysis of responses provided at the end of Q10; name of supermarket(s) preferred

over main supermarket and reason(s), the 276 cases were sorted according to spurious &

latent loyalty and cases with a construct score greater than 3.04 (3.04 being the median for

this construct) were selected. 138 cases exceeded this threshold but only 39 cases were used

in the analysis as 99 respondents had not answered this section. 87% of the respondents (34

out of 39) reported Keells or Arpico as their preferred grocery retailer; confirming the

preferred status enjoyed by these two retailers. 18 of the 20 respondents (90%) whose main

supermarket is Cargills mentioned Keells or Arpico as their preferred alternative. Interestingly,

6 of the 8 (75%) respondents shopping mainly at Keells mentioned Arpico as their preferred

choice. These findings lend further support to previous findings that Keells and Arpico are the

preferred choice for Sri Lankan respondents although Arpico appears to be the most preferred

retailer.

Respondents identifying Keells and/or Arpico as their preferred supermarket(s) referred to

better quality and customer service; “Quality of goods, high quality customer service when

compared to others” (Female: 55-64). The greater choice, variety and range of products

available at these two supermarkets were also identified by several respondents; “Wide

variety of products; One-Stop-Shop” (Female: 55-64). Being able to buy everything under one

roof was identified by several respondents who reported Arpico as their preferred

supermarket. The influence of customer service as identified herein is rather surprising. This is

because respondents in both countries placed high levels of importance on this element.

However, responses to this open-ended question shed further light on the possibility of

customer service having greater influence on the choice of retailer in Sri Lanka.

5.3.7.5 Overrepresentation of the 25-34 Age Group in Sri Lanka

To ascertain whether significant differences exist between the 25-34 age group and the other

age groups, Mann-Whitney U Tests were conducted using the construct scores from the CFA as

well as the 24 items used under Q7 (Store Image). Age (originally consistent of 7

subcategorises) was divided into 2 categories (1; Other age groups and 2; 25-34 age group). In

respect of the post-CFA constructs, a significant difference (p < .05) was only observed for

Aversion to Tracking (p = .016, r = 0.15) accompanied by a small effect. With regard to the

Store Image attributes, significant differences only exist for Customer Service and Reputation

(p = .043 and .028 respectively) with small effects (r = 0.12 and 0.13 respectively). A Chi-square

146

test for independence (with Yates Continuity Correlation) was conducted to establish whether

a significant difference exists between the 25-34 age group and the other age groups in

respect of supermarket usage (1- Low and 2- High). An insignificant difference X2 (1, n= 256), p

= .256, phi = .079 suggests that supermarket usage is not different across the two age

categories. While acknowledging the overrepresentation of this age group, it can be concluded

that the validity of the data is not affected by any sampling bias.

5.3.8 Summary and Confirmatory Framework Development

The confirmatory framework presented below summarises the key findings presented in this

section. Findings where the UK has a significantly greater value are in red whilst those

significantly greater in Sri Lanka are in blue. Components in the refined conceptual framework

not followed up by statistical tests (Instant Rewards and Special Occasions) have been

excluded. Nonsignificant differences; denoting similarities between the two countries are in

black.

Findings show that there is greater openness to soft benefits, tiered schemes, preferential

treatment and tracking in Sri Lanka while there is greater aversion to tracking and own labels

in the UK. There is greater overall true loyalty in the UK despite greater customer promiscuity

in terms of store usage. There is greater spurious & latent as well as captive loyalty in Sri

Lanka. There are high expectations of customer service and low aversion to fresh produce in

the two countries.

147

Figure 5.7: Confirmatory Framework * Whilst a significance difference is not reported, these two constructs are depicted as clear differences were identified from the qualitative findings- These will be discussed in greater detail in the Discussion chapter

Differences in Customer Loyalty Latent & Spurious Loyalty

p = <.001 r = .18 True Loyalty

p = <.001 r = .14 Captive Loyalty

p = .018 r = .1

Soft Benefits p = <.001

r = .62

Tracking p = <.001

r = .53

Customer Loyalty

Own Labels p = <.001 r = .76

Promiscuity p = <.001

Grocery Shopping

Habits

Specialist Stores* p = .827

Loyalty Programmes

Tiered Schemes p = <.001

r = .62

Preferential Treatment

p = <.001 r = .43


Store Image Customer Service; p = .547* Freshness; p = .046 Price; p = .085 Loyalty programmes; p = <.001 Quality; p = .048 Deals & Offers; p = .002 Convenience; p = <.001 Store Layout; p = <.001 Range & Choice; p = <.001 Ethical considerations; p = <.001

148

Chapter 6

Discussion

149

6.1 Introduction

This chapter discusses the key findings presented in the previous chapter in relation to the

existing literature and market data. Differences in customer loyalty to the different grocery

retailers in the UK and Sri Lanka are discussed, followed by a discussion of the overall degree

of customer loyalty in the two countries. Customer perceptions of loyalty programmes are

then discussed, after which, differences and similarities in grocery shopping habits as well as

the importance of customer service are discussed. Summative diagrams are presented at the

end of each of the abovementioned sections and these are combined at the end of this

chapter to summarise the key points discussed. This combination further aids the

development of the theoretical framework which ties the different findings together; providing

a holistic illustration of the influence of national culture and industry structure on grocery

retail customer loyalty.

6.2 Differences in Loyalty

The findings presented in Chapter 5 clearly show that national culture and industry structure

have an influence on grocery retail customer loyalty. However, there are differences in the

way customer loyalty is affected by these two factors in the two countries. Elements of

industry structure have a greater influence in the UK whilst elements of national culture exert

greater influence in Sri Lanka. Figure 6.1 depicts the situation in the UK where it can be seen

that customers demonstrate true loyalty; i.e. high repeat patronage and high relative attitude

(Dick and Basu 1994) towards the eight main grocery retailers. This is somewhat surprising due

to the highly competitive nature of the industry. It is also striking that despite having access to

a range of retailers, customers are in fact truly loyal to their main retailer. Similarly, low levels

of spurious loyalty; i.e. high repeat patronage and low relative attitude (Dick and Basu 1994)

reported towards these eight retailers further confirm this phenomenon.

Figure 6.1: Loyalty to Main Grocery Retailer in the UK*

True Loyalty Big Four, Discounters

& Premium

Latent Loyalty Premium

Spurious Loyalty -

No Loyalty -

High Low *Main Grocery Retailers in the UK

Big Four: Tesco, Sainsbury’s, Asda and Morrisons Premium: Waitrose and Marks & Spencer Discounters: Aldi and Lidl



Low

High

150

Favourable attitudes towards their respective main grocery retailers are strongly influenced by

positive perceptions of product quality. Although these eight grocery retailers have differing

price positions, customers perceive a common focus on product quality. Such a focus could be

influenced by these retailers understanding the importance of quality to customers alongside

price (Ranaweera and Neely 2003). This is reflected in their recent advertising campaigns

which focus on aspects such as provenance, farming, sustainability and overall quality.

Customer awareness and appreciation of these initiatives is clearly reflected in the primary

data. This finding corroborates previous research; Das (2014) that highlights the positive

impact of retailer perceived quality on customers’ store loyalty. The common quality focus

across the eight grocery retailers despite their varying pricing and positioning strategies

appears to influence positive customer attitudes; resulting in true loyalty to their main retailer.

It is also evident from Figure 6.1 that some customers demonstrate latent loyalty; i.e. high

relative attitude and low repeat patronage (Dick and Basu 1994) towards the two premium

retailers. This appears to be a result of their positioning strategies, where a premium image is

portrayed in their advertising; an approach not emulated by the other main retailers (Mintel

2016b). The impact of this positioning is reflected in customer perceptions of these retailers as

evidenced by the primary data. The plausibility of such a phenomenon is reported by Das

(2014) where brand personality which is used to position a brand, has been found to influence

customers’ store loyalty. The influence of industry structure is again evident in the way the

retailers position themselves; Marks & Spencer and Waitrose as premium, the Big Four as mid-

range and the two discounters as value-led. The effect of this positioning is somewhat

reflected in customer purchasing behaviour; the two premium retailers usually attracting

customers from higher socio-economic groups (Mintel 2016c). Latent loyalty to these two

premium retailers also appears to be a result of the greater perceived quality as highlighted by

the primary data; again resultant of their positioning strategies. The likelihood of such

perceptions is confirmed by Das (2014) who goes on to state that retailer personality also has a

positive impact on perceived retailer quality.

Waitrose and Marks & Spencer attracting latent loyalty from customers of other retailers can

be deemed positive for these two premium retailers. However, the primary data highlights

that latent loyalty towards the two premium retailers is a result of customers considering

these retailers to be too expensive and not offering adequate value for money. Lowering prices

may not be an alternative for these retailers due to their premium positioning. However, these

two retailers could induce greater repeat purchasing from latently loyal customers by offering

special deals and offers via their loyalty programmes. Waitrose currently offers various offers

through its ‘myWaitrose’ card and perhaps may be able to target latently loyal customers using

151

this approach. Further, the premium retailers may also be able to capitalise on customer

preferences during special occasions by offering incentives to obtain their patronage as

highlighted by the primary data.

Figure 6.2 depicts the situation in Sri Lanka where it is evident that customers demonstrate

true loyalty to Keells and Arpico; i.e. the two Better grocery retailers. This favourable position

enjoyed by these two retailers is further confirmed by the low degree of spurious loyalty

reported towards them. It is evident from the primary data that this is a result of greater

customer perceptions of quality and customer service. These perceptions suggest that these

two retailers may in fact be perceived as premium chains by customers. However, this is rather

surprising as these two retailers do not portray a premium image in their advertising; as do

Marks & Spencer and Waitrose in the UK. Therefore, such perceptions do not appear to be a

result of any positioning efforts undertaken by the retailers but rather, purely a result of

customer perceptions. It can also be seen from Figure 6.2 that customers demonstrate

spurious loyalty towards the market leader; Cargills as well as the other retailers. This shows

that whilst engaging in repeat patronage, customers do not have a high relative attitude

towards these retailers (Dick and Basu 1994). Spurious loyalty to these retailers also appears to

be a result of negative customer perceptions of quality and customer service as evidenced by

the primary data.

Figure 6.2: Loyalty to Main Grocery Retailer in Sri Lanka* It is also evident from the figure above that latent loyalty is demonstrated towards the two

Better retailers. Such loyalty is apparent from customers of the market leader; Cargills and the

other retailers. Spurious loyalty to these retailers means that customers in fact have latency

towards the two Better retailers. Despite engaging in repeat patronage, customers hold higher

relative attitudes towards the two Better retailers. As discussed above, greater perceptions of

quality and customer service appear to cause such behaviour as highlighted by the primary

data.

True Loyalty Better

Latent Loyalty Better

Spurious Loyalty Cargills & Other

No Loyalty -

Low High


*Main Grocery Retailers in Sri Lanka

Better: Keells and Arpico Cargills: Market leader Other: Laugfs, Sathosa


Low

High

152

Differences in customer loyalty as mentioned above are influenced by elements of industry

structure as well as elements of national culture. From a structural standpoint, retailer focus

appears to influence customer perceptions of quality and customer service. Whilst not

positioned as premium chains, the two Better retailers focus on quality and customer service

and this is confirmed by the primary data. The marker leader; Cargills on the other hand

focuses on an EDLP; everyday-low-price approach with its focus primarily on price (Cargills

Ceylon 2018). The primary data also highlights that low levels of quality are attributed to the

other retailers. As a result, low levels of perceived customer service and quality are clear

towards Cargills and the other retailers due to a lack of emphasis on these factors as evidenced

by the findings.

Given that Keells and Arpico do not portray a premium image (as done by Waitrose and Marks

& Spencer); they may be able to reduce prices to attract latently loyal customers. This could be

a sensible approach as the primary data identifies that these two retailers are considered

relatively expensive. Cargills needs to be cautious of the spurious nature of their customer

base. The retailer could overcome this potential pitfall by focusing on improving customer

service as well as freshness and quality of produce as highlighted by the primary data.

Customer emphasis on quality also appears to be influenced by the familiarity with and the

influence of traditional trading formats. With the country’s grocery sector dominated by

traditional retailing such as specialist stores and wet markets (LMRB 2016), supermarket

shoppers appear to compare the quality of fresh produce available in supermarkets against

those in more traditional markets. Due to their knowledge of the quality in more traditional

formats, customers may tend to prefer modern grocery retailers who offer similar standards of

quality. As evidenced by the primary data, such behaviour is likely in countries such as Sri

Lanka that have developing retail structures; Reardon et al. (2012) where the growing modern

retailing formats constitute a minority. This finding corroborates much earlier findings;

Goldman (1974) where customers in developing markets have been found to rely heavily on

traditional stores and use such stores as a benchmark for aspects such as quality. The need for

quality could also be a result of the country’s food culture where cooking is mainly done from

scratch; using raw ingredients (Albala 2010). This finding is likely to apply to other countries

with similar industry structures and food cultures. Overall, these elements of industry

structure and food culture appear to strongly influence the importance placed by customers

on the quality of products and fresh produce.

The emphasis placed on customer service and the preference for grocery retailers providing

superior customer service in Sri Lanka appear to be driven by elements of national culture.

153

More specifically, the high Power Distance (Hofstede Insights 2018) in the country can be

attributed to this. This corroborates previous research where customers from high Power

Distance countries have been found to expect high levels of customer service (Schmitt and Pan

1994). Such expectations are a result of customers focusing on social status (Hofstede et al.

2010) due to perceived status differences between customers and employees (Mattila 2000).

The clear hierarchical order in high Power Distance societies (Hofstede 1991) may mean that

customers naturally occupy a higher level compared to retail staff for instance; resulting in

such high service expectations. Such social hierarchies identify an individual’s place in society;

highlighting one’s social status so that due respect can be paid (de Mooij and Hofstede 2010).

Furthermore, the high Femininity in the country can also be taken to influence this

phenomenon. Due to their high Femininity (Hofstede Insights 2018), Sri Lankan customers may

be more prone to relationship building and open to such efforts from retailers. The likelihood

of this is evidenced by the primary data where friendly customer service was viewed positively

in Sri Lanka whilst done so negatively in the UK; which is likely due to the high Masculinity in

the UK (Hofstede et al. 2010; Hofstede Insights 2018).

6.3 Overall Loyalty in the Two Countries

It is evident from the primary data that customers are more truly loyal to their main grocery

retailer in the UK than in Sri Lanka. This is highlighted by high levels of true loyalty to the eight

main retailers supplemented by low levels of spurious loyalty. This finding shows that despite

using more stores per month, UK customers are in fact more truly loyal to their main retailer

compared to their Sri Lankan counterparts. Whilst supporting Luceri and Latusi’s (2012)

argument that a customer’s patronage set widens as the number of alternatives increases, this

finding clearly highlights that such behaviour does not negatively influence true loyalty to the

main retailer. Despite the two Better retailers enjoying true loyalty, the overall degree of true

loyalty in Sri Lanka is lower. This is due to the moderate levels of true loyalty and spurious

loyalty demonstrated towards the market leader; Cargills as well as Laugfs and Sathosa.

These findings clearly highlight that such differences are partially influenced by elements of

industry structure. In the absence of previous research, these findings provide novel insights in

this respect. Furthermore, these findings add to the limited research in the area of cross-

cultural grocery retail loyalty. However, findings from this study only provide limited support

to previous research findings as highlighted in Table 6.1. For instance, this study’s finding that

UK customers are more store-loyal lends support to the argument posited by Lam (2007). It is

arguable that UK customers, due to their high Individualism as per Lam (2007) are more loyal

to their main store. However, the other argument that high Uncertainty Avoidance leads to

154

high customer loyalty is questionable as there is a low level; 35 (Hofstede et al. 2010; Hofstede

Insights 2018) of Uncertainty Avoidance in the UK. Seock and Lin’s (2011) finding that high

Collectivism leads to greater customer loyalty tendencies is not supported as lower true loyalty

is reported in the highly Collectivist Sri Lanka. Similarly, Ndubisi et al.’s (2012) finding that high

Uncertainty Avoidance leads to greater customer loyalty is not supported due to greater true

loyalty in the UK, despite its low Uncertainty Avoidance culture.

Previous Research

Context Previous Findings Conclusions from Research Findings

Lam (2007) Brand loyal proneness in Australia

Customers with high Individualism (IDV) and high Uncertainty Avoidance (UAI) have a greater proneness to be brand loyal

Supports the finding relating to IDV but questions are raised about UAI as the UK has a low level of UAI (Hofstede et al. 2010; Hofstede Insights 2018)

Seock and Lin (2011)

Young customers’ loyalty towards apparel retail stores in the USA and Taiwan

Customers from countries with high Collectivism (COL) have greater loyalty tendencies

Contradicts the finding as the UK has a high IDV score; 89 (Hofstede et al. 2010; Hofstede Insights 2018)

Ndubisi et al. (2012)

The impact of Uncertainty Avoidance (UAI) on customer loyalty in the banking sector in Turkey and Malaysia

Turkish customers (high UAI) were found to be more loyal than Malaysian customers (low UAI). UAI has a positive impact on customer loyalty

Contradicts the finding as the UK and Sri Lanka have low UAI scores; 35/45 (Hofstede et al. 2010; Hofstede Insights 2018)

Table 6.1: Research Findings versus Previous Findings

Although it is clear that national culture does have an influence on customer loyalty, this

study’s findings suggest that the influence is indirect. This is contrary to the previous studies

where the impact is argued to be direct. The potential indirect influence of national culture is

for instance evident through the finding in Sri Lanka with regard to customer service. Previous

studies have used national culture in isolation whereas this study also uses industry structure.

The combined application of these factors highlights how national culture may also influence

industry structure. For instance, customer loyalty to the main store in Sri Lanka being linked to

customer service appears to be a result of the high Power Distance in the country.

Consequently, it is likely that industry structure could be influenced by this cultural element

where grocery retailers may need to focus on customer service; thus, resulting in a change in

the setup of the industry.

155

Figure 6.3 summarises the aspects discussed in the above sections. It is clear that national

culture does not have a direct influence on customer loyalty (as represented by the dashed

arrow). Instead, national culture influences customer loyalty through customer service in Sri

Lanka while such an influence is not evident in the UK. Customer service will have a positive

direct influence on customer loyalty as confirmed by previous research and this is depicted by

a solid arrow. It is also likely that customer emphasis on customer service may influence

retailer focus in Sri Lanka. This indirect influence is depicted by a dashed arrow. Elements of

industry structure on the other hand, directly influence customer loyalty in both countries. In

the UK, retailer focus and premium positioning have a direct influence whilst retailer focus and

industry development have an influence in Sri Lanka. These influences are depicted by solid

arrows.

Figure 6.3: Influence of national culture and industry structure on customer loyalty in the UK and Sri Lanka (Solid arrows depict direct influences whilst dashed arrows depict indirect influences)*

6.4 Loyalty Programmes 6.4.1 Attitudes towards Tracking

Clear differences between the two countries are evident through the primary data. UK

customers hold negative attitudes towards being tracked by loyalty programmes whilst their

Sri Lankan counterparts are open to such tactics. These differences could be influenced by

elements of national culture as well as aspects of industry structure. In terms of industry

structure, the lifecycle stage of loyalty programmes in the two countries may be useful in

explaining these differences. The pervasiveness of loyalty programmes in the UK may have led

to customers being more aware of data collection practices and holding negative attitudes as

evidenced by the primary data. On the contrary, the absence of tracking in Sri Lanka may have

led to participants identifying the need for their behaviour to be tracked. Such initiatives

therefore, may be more welcomed as they provide a novel experience to Sri Lankan

customers.

National Culture High PDI & Low MAS

Industry Structure Retailer focus & development (SL) Retailer focus (UK) Premium positioning (UK)

Customer Service

Customer Loyalty

*Key to Abbreviations PDI: Power Distance MAS: Masculinity SL: Sri Lanka

156

These differences also appear to be influenced by two dimensions of national culture;

Individualism and Masculinity. Aversion to tracking in the UK appears to be a result of the high

Individualism in the country (Hofstede Insights 2018), where customers value privacy and

personal space (Hofstede et al. 2010). This finding corroborates previous research where

customers with high privacy concerns have been found to be averse to loyalty programmes

that track and target customers (Enzmann and Schneider 2005). This previous finding can also

be used to explain the situation in Sri Lanka where customers appear to be open to such

initiatives due to their lack of emphasis on privacy and personal space due to their high

Collectivism (Hofstede et al. 2010; Hofstede Insights 2018). Sri Lankan customers may also

seek to maintain relationships and be persuaded by a retailer’s relationship building initiatives

(Laroche et al. 2004). This emphasis on relationship building can be attributed to the low

Masculinity in the country (Hofstede Insights 2018), where such customers are more likely to

prefer relationships and cooperation (Hofstede et al. 2010). On the other hand, UK customers’

aversion to be tracked could be linked to high Masculinity which does not value relationship

building and cooperation (Hofstede et al. 2010).

There is clear potential to track customer behaviour in Sri Lanka; providing retailers the

opportunity to develop relationships with customers via two-way communication (Meyer-

Waarden and Benavent 2009). However, UK retailers may find it more difficult to track

customer behaviour due to negative customer attitudes. Therefore, UK retailers will need to

ensure greater transparency as highlighted by the primary data so that customers are aware of

how their data is used. Further, privacy-aware programmes (Blanco-Justicia and Domingo-

Ferrer 2016) may also be appropriate; allowing customers to decide how much data they

share. Highlighting the benefits of sharing information may further alleviate customers’ privacy

concerns (Taylor et al. 2015; Leppäniemi et al. 2017). This is because customers weigh the risks

and benefits of participating in loyalty programmes (Norberg et al. 2007).

6.4.2 Rewards

Similarities between the two countries in respect of the types of preferred rewards are evident

from the primary data. Specifically, both countries’ customers demonstrate a preference for

instant or segregated (broken-up) rewards. This preference stems from a common lack of

willingness in the two countries to accumulate points for future use. National culture appears

to influence this preference for such rewards. Due to the low Long Term Orientation in both

countries (Hofstede Insights 2018), customers appear to focus on short term gains (Hofstede

et al. 2010). Similarly, customers from low Long Term Orientated cultures are known to expect

instant gratification of their material, social and emotional needs (Hofstede et al. 2010).

157

Therefore, it is likely that customers in the two countries expect instant gratification through

instant and segregated rewards, instead of having to wait longer for delayed or aggregated

rewards.

This finding chimes with previous research, Park et al. (2013), where less Long Term Oriented

customers were found to be positively affected by immediate rewards. Findings from other

studies; Jang and Mattila (2005) in similar cultures such as the US provide similar findings;

offering further support to findings from this study. However, this study’s finding questions the

argument that customers in countries with high Masculinity such as the UK prefer immediate

rewards as discussed by Mattison Thompson and Chmura (2015). This is because

interpretation of this argument indirectly suggests that customers in countries with high

Femininity such as Sri Lanka would prefer delayed rewards. However, findings from this study

show that Sri Lankan customers in fact, prefer immediate rewards; refuting the argument

presented by Mattison Thompson and Chmura (2015).

To be effective, retailers need to understand and offer the types of rewards that are perceived

favourably by customers (Jang and Mattila 2005; Omar et al. 2010). It is clear that instant or

segregated rewards would be more effective in countries with low Long Term Orientation

cultures. Such rewards have also been found to be more effective in low involvement contexts

such as grocery retailing as customers may perceive greater value in such rewards (Yi and Jeon

2003; Meyer-Waarden 2015). There is also evidence to suggest that instant reward

programmes in fact result in incremental shopping visits (Minnema et al. 2017). However,

customers who prefer instant or segregated rewards may be prone to switching. This is

because such rewards do not create switching costs and may not discourage customer

switching (Park et al. 2013) compared to aggregated rewards (Zhang et al. 2000). The positive

impact of such instant rewards may only be on behaviour as their impact on attitudes is

inconclusive. Therefore, grocery retailers need to evaluate how such strategies fit in with their

overall objectives. While offering instant rewards in the short term, grocery retailers in

countries such as the UK and Sri Lanka may need to encourage customers to accumulate

points for future use in order to create switching costs in the hope of discouraging customer

switching.

158

6.4.3 Soft Benefits, Preferential Treatment and Tiered Schemes

Clear differences between the two countries are evident through the primary data in respect

of soft benefits, preferential treatment and tiered schemes. Whilst Sri Lankan customers

demonstrate openness to these three attributes, the opposite is evident of UK customers.

Differences in elements of national culture can be attributed to these differences.

Openness to soft benefits such as greetings or gifts on special occasions in Sri Lanka appears to

be a result of the Collectivist and Feminine cultural orientations in the country (Hofstede

Insights 2018). Sri Lankan customers may not be as concerned about their privacy if retailers

attempt to offer soft benefits due to their Collectivist orientations. UK customers on the other

hand, may be averse to such initiatives as they may consider such initiatives to be intrusive.

This is due to the Individualist culture in the UK which values personal space (Hofstede et al.

2010). The high Femininity in Sri Lanka also appears to play a part where customers may be

open to such initiatives due to their emphasis on relationship building and cooperation

(Hofstede et al. 2010). This supports previous research; Laroche et al. (2004) where high

Femininity has been found to positively influence customer openness to retailers’ relationship

building efforts. On the other hand, UK customers could be averse or less open to such

initiatives due to their lack of emphasis on relationship building efforts (Hofstede et al. 2010).

Soft benefits have been found to enhance attitudinal loyalty as they create emotional

commitment (Gable et al. 2008; Vesel and Zabkar 2010). Such benefits may also be more

attractive to customers as they are more experiential in nature (Cedrola and Memmo 2010).

Although more difficult to practice, soft benefits are harder to imitate compared to hard

benefits (Gable et al. 2008). Grocery retailers can use soft benefits in Sri Lanka to improve

attitudinal preferences which could lead to greater true loyalty. Given the lack of openness to

such benefits in the UK, hard benefits that are more monetary oriented would be more

effective. While being easier to employ, hard benefits can also be easily imitated by

competitors (Gable et al. 2008). However, loyalty programmes should offer a mixture of hard

and soft benefits to customers (Omar et al. 2010) to ensure that varied customer requirements

are adequately met.

The need for preferential treatment and tiered schemes in Sri Lanka appears to be a result of

the high Power Distance in the country (Hofstede Insights 2018). Customers in such countries

have been found to focus on status (Hofstede et al. 2010) where customers would compare

their rewards against those received by others (Beck et al. 2015). Such behaviour is driven by

prestige and envy caused by preferential treatment and status (Steinhoff and Palmatier 2016).

However, previous research in the area provides mixed findings. Some studies (Lacey et al.

159

2007; Garcia-Gomez et al. 2012; Tanford 2013) suggest that preferential treatment and tiered

schemes have a positive influence on customer loyalty. Such initiatives, while positively

influencing customers in higher tiers may not alienate other customers as they may perceive

this to be just (Colliander et al. 2016).

However, preferential treatment and tiered schemes could also have a negative influence on

customer satisfaction and loyalty. This is due to concerns of favouritism and fairness amongst

customers receiving preferential treatment as well as those who do not (Xia and Kukar-Kinney

2014). This could be worse in Sri Lanka due to the greater focus on status (Hofstede et al.

2010). Customers who do not receive preferential treatment may dislike such practices which

could lead to customer dissatisfaction (Wang and Lalwani 2013). This may also be due to envy

caused by such treatment and status (Steinhoff and Palmatier 2016). Furthermore, grocery

retailers need to be aware that not all customers are the same in respect of such treatment as

previous research suggests that customers in fact, may have varied perceptions of such

treatment. For instance, Butori and De Bruyn (2013) suggest that such treatment, while well

received by some, could anger or embarrass others.

Such initiatives could also be ineffective if customers’ own perceptions of their status or

eligibility are different to those set by the retailer (Pez et al. 2017). Such a ‘targeting mismatch’

could lead to customer dissatisfaction and disloyalty. Grocery retailers also need to consider

the potential negative impact of demotions when customers are demoted to lower tiers based

on set criteria. This is because demotions could have a negative influence on attitudes and

behaviour (Ramaseshan and Ouschan 2017). Whilst tiered schemes and preferential treatment

are more prevalent in industries such as airlines and hotels, the suitability of such practices in

low involvement industries such as grocery retailing has been questioned. This is because

structured schemes may not necessarily create feelings of status in such industries (Arbore and

Estes 2013).

Whilst preferential treatment and tiered schemes may be appropriate in countries such as Sri

Lanka, such initiatives need to be used in a subtle manner. A good example of this is the

‘Sparks’ loyalty programme used by Marks & Spencer where no explicit tiers are

communicated. However, customers are entitled to different types of preferential treatment

based on their accumulation of points. Emulating such an approach, grocery retailers in Sri

Lanka could reward customers based on their loyalty status by offering differential rewards

through individual communications to avoid alienating other customers who may not be

entitled to such rewards. Further, grocery retailers could also make use of soft benefits to

reward customers differently based on their loyalty status.

160

Figure 6.4 illustrates the influence of these cultural characteristics. The low Long Term

orientation in both countries appears to make customers prefer instant rewards. High

Collectivism and Femininity in Sri Lanka appear to create openness to soft benefits while high

Power Distance makes Sri Lankan customers expect preferential treatment and tiered

schemes. Familiarity with loyalty programmes and subsequent experience of being tracked

may have made UK customers more averse to tracking while the opposite is evident in Sri

Lanka, perhaps due to the novelty and lack of such practices. Whilst high Collectivism in Sri

Lanka may make customers open to tracking, high Individualism in the UK may cause aversion.

The influence of loyalty programmes on customer loyalty has been confirmed by previous

research; Meyer-Waarden (2008). These findings show that customer perceptions of loyalty

programme attributes are influenced by cultural and structural characteristics, influencing

customer preferences for loyalty programmes; thereby impacting customer loyalty.

Figure 6.4: The influence of national culture and industry structure on customer perceptions of loyalty programme attributes

6.5 Grocery Shopping Habits 6.5.1 Aversion to Own Labels and Influence on Customer Loyalty

Stark differences in customer attitudes and behaviour are evident in the two countries

towards own label products. While Sri Lankan customers are open to own labels, UK customers

demonstrate moderate aversion as evidenced by the primary data. This finding is interesting in

both countries due to the respective lifecycle stages of these products. The low aversion in Sri

Lanka is interesting due to the relative novelty of such products and the lack of satisfaction

reported towards them by focus group participants. The high levels of customer satisfaction

with these products as evidenced by the primary data are rather striking due to the lack of

own label penetration in the country. This is because low customer perceptions of quality have

been previously reported in countries with low levels of own label penetration (Shannon and

National Culture Low Long Term

Orientation- UK & SL

Instant Rewards

Soft Benefits

National Culture High Collectivism & High Femininity- SL

Preferential Treatment

Tiered Schemes

Customer Loyalty

Tracking

National Culture High Power Distance- SL

Industry Structure Familiarity- UK

Unfamiliarity- SL

161

Mandhachitara 2005). This finding questions other previous research as well; de Mooij (2004)

where Asian customers have been reported to avoid own labels and prefer national brands.

However, this study was conducted amongst customers living in urban or suburban areas with

greater access to modern food retailing formats. It is likely that such customers have become

accustomed to own labels due to prolonged exposure. It is therefore possible that customers

in less structurally developed parts of the country may be more averse to own labels. This is

because they rely on more traditional formats and are less familiar with such products given

the lack of market development (Budhathoki 2014). However, further research is required to

investigate this in greater detail.

The moderate aversion in the UK is rather surprising given the high market share enjoyed by

own-labels and the high levels of satisfaction (Key Note 2015). Despite own labels in the UK

being forecasted to gain more market share in the next 5 years (Key Note 2015), customers

appear to be cautious of these products. This is again striking as future growth has been

attributed to increasing perceptions of quality parity with national brands (IGD Shopper Vista

2017b) and increasing levels of customer satisfaction (Key Note 2015). Moderate levels of

aversion could perhaps be explained by the tiered nature of these products (Key Note 2015).

While value and mid-tier products cater to price oriented customers, the premium ranges

focus on premium quality. Premium labels compete with national brands and are positioned

on par with such brands in relation to quality. It is plausible that customer preferences are thus

influenced by the tiered nature of these products. This is further supported by customer

preferences for premium own labels (Mintel 2017b) where customers may be open to

premium labels whilst being averse to value and mid-tier labels. Latency towards national

brands has also been reported in the UK (Key Note 2015) despite the high market share of own

labels. It is likely that while customers may perceive premium own labels in a positive light,

they may perceive value and mid-range products to be inferior in relation to national brands.

Since the data from this study does not serve to substantiate this speculation, more focused

research should be carried out to ascertain whether the tiered nature of own labels

contributes to this phenomenon.

An influence of national culture is not evident although some reference can be made to

previous studies. Budhathoki’s (2014) finding that high Individualism and Masculinity have a

positive impact on own label performance is corroborated by this research. The high market

share enjoyed by these products in the UK (Key Note 2015) reflects the possible influence

national culture may have on own label performance. However, this research focused on

perceptions of own-labels and it is difficult to link any elements of national culture to the

identified differences. On the other hand, industry structure appears to be more influential in

162

this respect and structural differences appear to influence customer perceptions in the two

countries. In the UK, tiered own-label ranges appear to influence customer perceptions;

resulting in moderate levels of aversion in general. Such a tiered structure is non-existent in Sri

Lanka and openness to such products in the country may only be representative of customers

living in relatively developed parts of the country. While previous research suggests that there

is a positive relationship between retail market development and own label performance

(Budhathoki 2014), this study’s findings do not suggest a straightforward influence of retail

market development on customer perceptions. This may be because the tiered nature; an

indication of a highly developed structure in the UK negatively influences customer

perceptions. In Sri Lanka, greater exposure to un-tiered own labels in developed parts of the

country appears to influence perceptions positively. The opposite is likely in less developed

areas where customers depend on traditional stores. It is also likely that the introduction of

tiered labels in Sri Lanka as the industry develops could result in similar differing perceptions

towards own-labels.

There is greater potential for product development in the less developed Sri Lankan own labels

market where retailers could start by introducing more own label product categories. This is

due to the limited number of product categories currently being offered by grocery retailers.

There is also potential to introduce a tiered structure as customer demand for own labels

increases. In the UK, grocery retailers need to focus more on premium own labels as customers

have been found to prefer such products if they are to forego national brands (Mintel 2017b).

An increased focus on premium own labels is evident as discounters such as Aldi and Lidl have

also started offering premium own labels (Mintel 2017b).

6.5.2 Special Occasions

Clear differences are evident between the two countries with regard to grocery shopping

behaviour on special occasions. UK customers tend to alter their choice of store for special

occasions, preferring premium grocery retailers on such occasions. On the contrary, Sri Lankan

customers are not influenced by such occasions. This may in part be due to the narrower range

of alternatives in the Sri Lankan market as well as differences in retailer promotional

strategies. While special occasions such as Christmas, Mother’s Day and Easter are heavily

promoted by UK grocery retailers to drive sales, Sri Lankan grocery retailers do not engage in

such actions. Whilst the influence of national culture is not evident, two elements of industry

structure can be attributed to these differences. The promotional emphasis evident in the UK

appears to influence customer behaviour on special occasions. Likewise, retailer positioning

also appears to have a similar impact.

163

The emphasis on special occasions in the UK has been previously reported (IGD Retail Analysis

2013) and is highlighted by advertising campaigns run by the eight main UK grocery retailers

(The Telegraph 2016). As discussed previously in this chapter, retailer positioning may lead to

customers preferring premium retailers. This has been identified by IGD Shopper Vista (2016)

where customers are reported to have a tendency to shop at premium grocery retailers for

Christmas. Such behaviour could be linked to the greater perceived image and quality attached

to such retailers. The use of premium retailers on special occasions could serve as a means of

personal indulgence or treating family members. On the contrary, such behaviour may not be

as prevalent in Sri Lanka due to the lack of premium grocery retailers such as Waitrose and

Marks & Spencer.

Findings from this study only suggest that customers may move towards more premium

grocery retailers on such occasions. However, previous research; IGD Retail Analysis (2015)

suggests that customers may also move towards discounters such as Aldi and Lidl. This is

because customers may shop at various retailers to spread their spending during expensive

times of the year (International Business Times 2015). Customers have also been found to

prefer specialist stores for special occasions (Mintel 2017c). Primary data suggests that such

preferences could be linked to greater perceptions of quality and value. Whilst unlikely in Sri

Lanka, such behaviour could affect customer patronage behaviour in the UK. Whilst

preferences may alter patronage behaviour on special occasions, it is unlikely that such

occasional shifts in customer patronage will result in longer term changes in loyalty to one’s

main store. Besides, it appears that UK grocery retailers have identified this possible threat

with the eight main retailers focusing on a more premium image with their advertising during

special occasions.

6.5.3 Aversion to Fresh Produce

Low levels of aversion to supermarket fresh produce are clear from the primary data in the

two countries. The finding in the UK is perhaps unsurprising as the majority of customers;

more than 65% are reported to buy fresh produce from grocery retailers (Mintel 2016d). It is

likely that customer familiarity with fresh produce at grocery retailers has resulted in low levels

of aversion. However, the finding is somewhat surprising in Sri Lanka due to the greater

prevalence of traditional formats and the developing nature of modern retailing (LMRB 2016).

It is also striking that customers demonstrate high levels of satisfaction towards fresh produce

available in modern grocery retailers. This finding questions previous research that suggests

otherwise. Seminal work in the area; Goldman (1974) suggests that customers in developing

countries would be more averse due to their dependence on traditional specialist stores. More

164

recent research; Reardon et al. (2012) supports this argument where customers are argued to

be less accustomed to buying fresh produce in modern retail formats. However, findings from

this research clearly show that such behaviour may have changed and that customers, at least

in urban areas of Sri Lanka are not averse to buying fresh produce from modern retailers. This

could be a result of increasing exposure to modern retailing formats coupled with traditional

stores being replaced by modern retailers in urban areas. However, this argument contradicts

previous studies in structurally similar India; Khare et al. (2014) and Pandey et al. (2015) where

cosmopolitanism, which represents modern retailing, is not seen to influence customer

behaviour towards traditional stores.

A common structural influence is evident in the two countries where familiarity with fresh

produce in grocery retailers appears to reduce levels of customer aversion. This appears to be

widely applicable to the UK due to its developed retail structure. Similarly, the same could be

said of customers in parts of Sri Lanka who are exposed to modern retailing; as represented in

the sample in this study. Therefore, it would be interesting to further explore whether

customers in more rural areas of the country with limited access to modern food retailers

would in fact, report greater levels of aversion. Despite low levels of aversion and high levels of

satisfaction in the two countries, customers demonstrate attitudinal preferences towards

specialist stores. It is clear that whilst preferring to shop at specialists, customers in both

countries use grocery retailers due to familiarity, convenience and price. It is therefore

arguable that customers have a certain degree of latent loyalty to specialist stores and a

similar degree of spurious loyalty to their grocery retailers in respect of fresh produce. This is

evidenced by the primary data where customers in both countries attribute greater quality and

value to such stores. However, this finding contradicts customer beliefs in the UK as reported

in previous research. Mintel (2016c) for instance, reports that customers in fact, believe that

grocery retailers offer greater value due to lower prices.

This discussion highlights that these three aspects of grocery shopping behaviour are

influenced by elements of industry structure in both countries. The figure below summarises

this and depicts how industry structure influences customer loyalty. It is clear from the primary

data that aversion to own labels and aversion to fresh produce both could have a negative

impact on customer loyalty to the main store in both countries. Although not tested

statistically due to unsatisfactory factor loadings in the two countries’ exploratory factor

analyses, it is also arguable that special occasions could have a negative influence on

patronage behaviour towards the main store in the UK.

165

Figure 6.5: The influence of industry structure on customer loyalty through grocery shopping habits

6.6 Importance of Customer Service

The primary data shows that customers in both countries place a high degree of importance on

customer service. High levels of importance are possible in Sri Lanka due to elements of

national culture. This could be explained by the high Power Distance in the country (Hofstede

Insights 2018), where customers may expect greater levels of customer service from staff as a

result of their focus on social status (Hofstede et al. 2010). Such behaviour could also be driven

by perceived status differences between customers and employees (Mattila 2000). Further

support is provided by previous research; Schmitt and Pan (1994) where the service-oriented

nature of Asian cultures has been identified to create high service expectations. The finding in

the UK on the other hand is interesting from a cultural viewpoint. This is because the UK has a

low Power Distance (Hofstede et al. 2010; Hofstede Insights 2018) and customers can be

assumed to place relatively lower importance on customer service. However, the finding

suggests otherwise and confirms previous research; Donthu and Yoo (1988) and Kueh and

Voon (2007) where customers from low Power Distance cultures would also have equally high

expectations of quality.

It should be noted that this study did not explicitly measure customer service expectations by

applying the SERVQUAL scale (Parasuraman et al. 1985) or similar such tool as done in these

previous studies. Instead, this study focused on the importance placed on customer service at

the macro level. Nevertheless, findings highlight that customers in the two countries consider

customer service to be an important factor regardless of their cultural differences; a finding

mirroring previous research where service expectations are largely indifferent across cultures

(Dash et al. 2009). However, it is interesting that whilst considered an important aspect in both

countries, Sri Lankan customers appear to follow-up on this in practice; more so compared to

Indu

stry

Str

uctu

re

Special Occasions

Specialist Stores

Customer Loyalty

Own Labels

166

their UK counterparts. This is highlighted by the influence of customer service in determining

true loyalty to the main retailer as discussed previously.

6.7 Chapter Summary and Theoretical Framework

Differences between the two countries in respect of national culture and industry structure

were discussed in this chapter with reference to the existing literature. Figure 6.6 below serves

as a summative theoretical framework, illustrating the key aspects discussed in this chapter.

These key elements were discussed in detail within sections 6.2 to 6.6 with figures 6.1 to 6.5

providing diagrammatic summaries. The legend next to the figure briefly describes the

elements and the discussed relationships. Customer loyalty is depicted as a holistic element for

clarity and simplicity. The various impacts on Dick and Basu’s (1994) customer loyalty

typologies have been discussed in earlier sections.

Sections 6.2 and 6.3 discussed how industry structure and national culture influence different

types of customer loyalty based on Dick and Basu’s (1994) typologies. In the UK, overall retailer

focus on quality appears to induce true loyalty towards the eight main grocery retailers.

Further, it is also evident that the premium positioning of Waitrose and Marks & Spencer

results in latent loyalty towards these two premium retailers. In Sri Lanka, greater focus on

product quality and customer service by Keells and Arpico was identified as creating true

loyalty towards these two retailers from their customers as well as latent loyalty from

customers of the other retailers. Conversely, a lack of emphasis on quality and customer

service can be attributed to lower true loyalty and greater spurious loyalty towards Cargills as

well as Laugfs and Sathosa. Sri Lankan customers’ focus on product quality could be alluded to

the degree of development in the industry where customers appear to compare the quality of

fresh produce, in particular, against those available in more traditional formats due to their

familiarity with traditional retailers. The section on the left in Figure 6.6 depicts this where

retailer focus, positioning and industry development constitute industry structure; influencing

customer loyalty in the two countries.

The section at the bottom of Figure 6.6 depicts the indirect influence of national culture on

customer loyalty through customer service as discussed in sections 6.2, 6.3 and 6.6. High

Power Distance and low Masculinity appear to influence greater customer expectations of

customer service in Sri Lanka. This indirect influence is depicted in Figure 6.3 where these

cultural traits influence customer loyalty through customer service. Further, these traits may

also influence the structure within the country where retailers such as Keells and Arpico

appear to have identified high customer expectations of customer service. The solid arrow

167

depicts the influence of customer service on customer loyalty as established by previous

research. The two dashed arrows depict the discussed indirect influence of national culture on

customer loyalty via customer service and the same indirect influence via industry structure in

Sri Lanka.

Section 6.4 discussed the greater influence of national culture on customers’ perceptions of

loyalty programme attributes and the limited influence of industry structure on the same. The

section at the top of Figure 6.6 depicts this. Low Long Term Orientation in both countries

appears to make customers prefer instant and segregated rewards. Whilst high Individualism

could be argued to make UK customers averse to being tracked, high Collectivism and

Femininity in Sri Lanka appear to make customers open to tracking. The lifecycle stage of

loyalty programmes also explains how industry structure influences this phenomenon. Whilst

UK customers may be familiar with tracking; influencing aversion, Sri Lankan customers’

unfamiliarity due to the lack of tracking appears to make them more open to such initiatives.

Finally, high Power Distance in Sri Lanka can be attributed to customers’ preference for soft

benefits, preferential treatment and tiered schemes.

The section on the right of Figure 6.6 depicts the discussion in section 6.5 which identifies the

influence of own labels, attitudes towards fresh produce and special occasions. Moderate

aversion to own labels in the UK appears to be a result of the tiered nature of these products

where customers have been found to prefer premium labels. Openness to these products in Sri

Lanka could be prevalent amongst customers in urban and suburban areas of the country due

to greater familiarity. Acceptance of fresh produce offered by retailers in both countries could

be alluded to customers’ familiarity as well as factors such as lower price and greater

convenience offered by organised grocery retailers. Finally, special occasions appear to

influence customer patronage behaviour in the UK where they prefer premium retailers.

However, such shifts in patronage could be considered transitory, given the high levels of true

loyalty in the country.

168

Figure 6.6: Theoretical Framework (Solid arrows depict direct influences whilst dashed arrows depict indirect influences)

National Culture Sections 6.2, 6.3 & 6.6

MAS PDI

Customer Loyalty*

Indu

stry

Str

uctu

re

Sect

ions

6.2

& 6

.3

Positioning

Industry Development

Loyalty Programmes Retailer

Focus

National Culture Section 6.4

LTO IDV

Specialist Stores

MAS PDI

Own Labels

Rewards Benefits Preference

Industry Structure Section 6.4

Tiers Tracking

Special Occasions

Customer Service

LTO- Long Term Orientation IDV- Individualism MAS- Masculinity PDI- Power Distance Low LTO in the UK & SL: Preference for Instant Rewards Low IDV/MAS in SL: Preference for Soft Benefits & open to Tracking High PDI in SL: Preference for Preferential Treatment & Tiered Schemes High IDV/MAS in the UK: Aversion to Soft Benefits & Tracking Familiarity in the UK: Aversion to Tracking Unfamiliarity in SL: Open to Tracking Retailer focus & positioning in the UK / retailer focus & industry development- SL influence customer loyalty to the main store High PDI & Low MAS influence customer loyalty via customer service in SL Aversion to own labels and special occasions (UK only): Negative impact on customer loyalty in both countries * Customer Loyalty is a holistic depiction of Dick and Basu’s (1994) four typologies as discussed in the commentary of this chapter

Indu

stry

Str

uctu

re

Sect

ion

6.5

169

Chapter 7

Conclusion

170

7.1 Introduction

As developed Western European markets remain stagnant, grocery retailers have expanded

into new markets; often those with developing sectors. The current global political climate has

brought about considerable human migration into developed Western European countries;

notably to the United Kingdom, France and Germany. Expansion into new markets creates new

challenges for international retailers with differences in national culture and industry structure

being two key aspects to consider. The influx of migrants into Western Europe also results in

diverse customer segments with different national cultural traits. Therefore, it is crucial for

grocery retailers to understand how customer loyalty is influenced by national culture and

industry structure (Laaksonen 1993; Straughan and Albers-Miller 2001; Khare 2012). Customer

loyalty has long been identified as a crucial requirement for business survival and growth

(Reichheld et al. 2000; Nguyen and Mutum 2012); even in grocery retailing. As a result, grocery

retailers in developed as well as developing markets have implemented loyalty programmes in

their efforts to generate and maintain customer loyalty.

Very little focused research has been carried out in the past to study the influence of

differences in national culture and industry structure on customer loyalty to grocery retailers.

Similarly, the impact of such differences on loyalty programmes remains thinly explored. The

few studies focusing on the influence of cultural differences on customer loyalty (Seock and Lin

2011; Khare et al. 2014; Grosso and Castaldo 2015; Pandey et al. 2015) and loyalty

programmes (Park et al. 2013; Hu and Weber 2014) remain fragmented. Furthermore, studies

attempting to combine both cultural and structural aspects to provide a holistic assessment of

their impact do not exist. This study provides a timely investigation of the above mentioned

phenomena by comparing two culturally and structurally different countries; the UK with a

developed grocery retail sector and Sri Lanka with a developing sector. These two countries

demonstrate differences and similarities based on Hofstede et al.’s (2010) dimensions of

national culture; providing a sound comparative base for this study.

The United Kingdom has experienced increasing rates of migration in recent years (Office for

National Statistics 2017c); resulting in more diverse customers in its domestic market (Guzi et

al. 2015). UK grocery retailers such as Tesco have expanded into developing South Asian

countries such as India (IGD Retail Analysis 2017c) and Pakistan through a potential joint

venture (The Express Tribune 2017). Despite such expansions, the impact of cultural and

structural elements has not been explored. Sri Lanka presents a suitable study setting given its

cultural and structural similarities with its South Asian neighbours. Sri Lanka therefore can be

considered a gateway market into the South Asian subcontinent. The island nation itself has

171

been identified as a country with significant potential for retail expansion. This is due to its

growing per capita GDP, favourable government policies towards foreign direct investments

and relative ease of doing business (Euromonitor International 2014; AT Kearney 2016). The

country has also been ranked among the top 15 countries for retail expansion, potential and

opportunities (AT Kearney 2017). Finally, this study conducting inductive research as a starting

point ensures that findings are in fact, reflective of customers’ actual thoughts and behaviour;

not governed by preconceived notions of culture or structure. This is crucial, especially in the

absence of adequate previous knowledge. This approach resulted in the generation of

insightful new knowledge which enabled the most pertinent qualitative findings to be used in

the development of the questionnaire.

This study provides novel and insightful findings; addressing academic recommendations for

research in this area and providing valuable recommendations for grocery retailers. The

previous research suggests that Hofstede’s dimensions of national culture have a direct

influence on customer loyalty in various contexts. This study in contrast, shows these

dimensions in fact, exert only an indirect influence. This study also questions the various

relationships between these dimensions and customer loyalty as presented in previous

studies, as they do not appear to be straightforward. In respect of loyalty programmes, this

study highlights the influence national culture has on customers’ perceptions of various loyalty

programme attributes. In so doing, the findings provide the first focused insight; national

culture does influence the role of loyalty programmes. In a similar manner, the role played by

industry structure is highlighted in this study. Findings show that structural differences have

both a direct and indirect influence on customer loyalty and influence perceptions of loyalty

programmes attributes.

7.2 Revisiting the Research Aim & Objectives

The aim of this research was to critically evaluate the influence of national culture and industry

structure on grocery retail customer loyalty in the United Kingdom and Sri Lanka. This aim has

been met by addressing the research objectives and this section revisits these objectives.

Objective 1: Develop a conceptual framework of grocery retail customer loyalty, national

culture and industry structure. The conceptual framework presented in Figure 3.8 (page 74)

draws together the existing literature and market data on customer loyalty, national culture

and industry structure; highlighting the lack of knowledge in this area. The identification of the

key antecedents of customer loyalty laid the foundation for the ensuing qualitative phase;

highlighting the key components to be explored in the focus group discussions. The

172

identification of the key characteristics of national culture and industry structure further aided

the qualitative phase by enabling the researcher to focus on particular topics of interest and

further explore these in the two countries’ discussions.

Objective 2: Refine the conceptual framework by exploring the impact of national culture

and industry structure on grocery retail customer loyalty in the UK and Sri Lanka. The refined

conceptual framework presented in Figure 5.6 (page 130) builds on the core components in

the preceding conceptual framework. The qualitative findings provide unique insights of the

influence of national culture and industry structure on the antecedents of customer loyalty as

identified within the conceptual framework. These findings provide further novel insights by

identifying how aspects such as grocery shopping habits and customer perceptions of grocery

retailers are influenced by national culture and industry structure. Given the lack of adequate

prior knowledge, this refined framework offers the first focused overview of the phenomenon

under study by enabling a preliminary interpretation of the findings in relation to the key

characteristics of national culture and industry structure.

Objective 3: Develop a confirmatory framework by further investigating the impact of

national culture and industry structure on grocery retail customer loyalty in the two

countries. The key qualitative findings presented in the refined conceptual framework aided

the development of the questionnaire used to address this objective. The rigorous statistical

analysis procedures followed confirm the reliability and validity of the survey items;

highlighting their suitability in future research. Figure 5.7 (page 147) summarises the key

findings confirmed and generalised by the statistical analyses. This framework confirms the

influence of national culture and industry structure on customer loyalty as previously

interpreted through the qualitative findings.

Objective 4: Present a theoretical framework highlighting the impact of national culture and

industry structure on grocery retail customer loyalty by synthesising the primary and

secondary data. The discussion of the key findings from the primary data in relation to the

existing literature and market data has shed greater light on how customer loyalty is

influenced by national culture and industry structure. This has been achieved by identifying

how customer loyalty is either directly or indirectly influenced by the key characteristics of

national culture and industry structure. The resulting theoretical framework presented in

Figure 6.6 (page 168) brings together the key findings and the key characteristics of national

culture and industry structure; providing the first such summative illustration of this

phenomenon.

173

Objective 5: Make recommendations to grocery retailers on how to enhance customer

loyalty in the UK and Sri Lanka. The final research objective is met through the discussion

within Chapter 6 which provides recommendations to grocery retailers on how to enhance

customer loyalty in the UK and Sri Lanka. Further recommendations are offered in this chapter

under section 7.5 where more focused recommendations are provided through the

identification of practical implications. In light of the lack of prior knowledge, these

recommendations would act as a starting point for both domestic and international grocery

retailers as they strive to enhance the loyalty of diverse customers in markets with different

cultural and structural characteristics.

7.3 Summary of Key Findings

Overall, the direct as well as indirect influence of national culture and industry structure on

customer loyalty warrants the argument that creation of customer loyalty, or in contrast

promiscuity, is a rather complex process. The likelihood of this is highlighted by the following

summary of this study’s key findings.

UK customers are more truly loyal to their main grocery retailer and this is a result of the eight

main grocery retailers’ common focus on quality. In an era where retailers pay increasingly

more attention to aspects such as customer experience and technology, the pivotal role played

by a relatively basic aspect such as quality is clear. Latency towards the two premium retailers

is evident and this is a result of positioning strategies used by these two retailers. Overall,

these differences in loyalty are influenced by industry structure; more specifically, retailer

focus on quality and premium positioning.

On the contrary, Sri Lankan customers are less truly loyal to their main grocery retailer with

spurious loyalty reported towards the market leader in particular. Customers however, are

truly loyal to Keells and Arpico and these two retailers also enjoy latency from customers of

the other retailers. This appears to be driven by differences in retailer focus and the greater

need for customer service. Despite both countries’ customers considering customer service to

be important, Sri Lankan customers’ true loyalty appears to be influenced more by this

element; attributable to their high Power Distance culture. Further, it is evident that the

greater need for freshness and quality influences customer preferences towards Keells and

Arpico over the others. The level of industry development in the country is apparent here;

customers comparing the quality and freshness of products available in organised grocery

retailing formats against that found in more traditional formats. This could be linked to the

174

predominance of traditional formats in the country and customers’ perceived greater

standards in such formats.

Sri Lankan customers are more open to soft benefits, preferential treatment and tiered loyalty

schemes compared to UK customers. These differences are influenced by their respective

national cultures. Greater Power Distance and Collectivism in Sri Lanka appear to make them

prefer such attributes due to their focus on status as well as openness to retailers’ relationship

building efforts. On the contrary, UK customers; due to their opposite national cultural traits

would be less open to such initiatives.

Aversion to being tracked in the UK is resultant of their national culture; more specifically their

high Individualism and high Masculinity. These can be argued to make them focus on privacy

and be less open to retailers’ relationship building efforts. Customer knowledge of being

tracked in the UK could also influence their aversions. This could be linked to the greater

structural development in the country where loyalty programmes are used to track customer

behaviour. Openness to tracking in Sri Lanka could be a result of high Collectivism and

Femininity where customers would be open to retailers’ relationship building efforts and value

relationships. The absence of tracking in the country due to the immaturity of loyalty

programmes may further make customers more open to being tracked.

Openness to instant rewards in the UK and Sri Lanka could be a result of the low Long Term

Orientation culture in both countries. As a result, customers would prefer instant or

segregated rewards due to their proneness to short term gains; driven by the need for instant

gratification of their consumption needs.

Moderate levels of aversion towards own labels in the UK could be a result of the tiered nature

of these products. As a result, customers appear to prefer premium labels; comparing them

against national brands. On the contrary, openness to own labels in Sri Lanka could be a result

of customers, at least in developed parts of the country being familiar with such products.

However, it is arguable that customers in less structurally developed parts of the country may

be averse to own labels.

Although not evident in Sri Lanka, special occasions may trigger changes in customer

patronage behaviour in the UK. Such changes could be influenced by customer preferences for

premium grocery retailers due to their premium image. However, such changes appear to be

transitory and do not necessarily influence the overall true loyalty to their main grocery

retailer.

175

Openness to fresh produce in organised grocery retail formats is clear in both countries. Whilst

this is more likely in the UK due to the dominance of such formats, openness in Sri Lanka is

somewhat surprising due to the dominance of traditional formats. However, customers in both

countries appear to hold a preference for traditional formats due to perceptions of greater

quality and value for money. Therefore, it can be argued that some degree of latent loyalty

exists towards traditional specialist retailers such as butchers and fishmongers in the two

countries.

7.4 Theoretical & Methodological Contributions

In light of the lack of prior research, this study is the first to critically investigate the influence

of national culture and industry structure on customer loyalty in the grocery retail context.

Whilst challenging the limited prior research, this study offers several new insights; making a

significant contribution to knowledge in the following ways.

This study’s first theoretical contribution is the enhancement of the limited existing knowledge

surrounding the influence of national culture on grocery retail customer loyalty. Findings

highlight an indirect influence of national culture where customer loyalty is influenced via

aspects such as customer service and loyalty programmes. This finding challenges existing

theory (Lam 2007; Seock and Lin 2011; Ndubisi et al. 2012) that suggests a direct influence. As

opposed to these previous findings, this study’s findings show that national culture is not a

direct influencer but, one that appears to moderate the relationship between customer loyalty

and the various contributing factors. In so doing, this study highlights that the influence of

national culture on customer loyalty is not as simplistic or linear, as argued by the previous

studies. The theoretical framework (Figure 6.6 on page 168) clearly depicts the moderating

role played by national culture and highlights that the influence of national culture is more

complexed than previously argued.

This study makes its second contribution by identifying the influence of industry structure on

customer loyalty. Findings show that differences in industry structure have both a direct and

indirect influence in both countries. For instance, aspects such as retailer focus and positioning

in the UK have a direct influence on differences in customer loyalty to the main retailer. The

indirect influence of structural differences is highlighted by the level of loyalty programme

maturity in the two countries; influencing customer perceptions of tracking in the two

countries. Overall, this hybrid approach provides a more holistic view compared to the

previous studies that only consider national culture. For example; Lam (2007) and Seock and

Lin (2011) found that only 21% and 15% of the variance in customer loyalty respectively, was

176

accounted for by national culture. The inclusion of another country-specific factor such as

industry structure highlights how differences in customer loyalty could be better understood.

Further, the inclusion of industry structure addresses the need for more focused research in

the context of retail internationalisation, as identified by Laaksonen (1993) and Chan et al.

(2011).

The third contribution of this study is its identification of how loyalty programmes are

influenced by differences in national culture and industry structure. Findings highlight that

national culture has a relatively greater influence compared to industry structure. Whilst

supporting the limited extant research (Park et al. 2013), findings provide additional insights

that can be used in future research to further investigate and confirm the influence of national

culture and industry structure. For instance, greater customer preferences for soft benefits,

preferential treatment and tiered schemes as identified in Sri Lanka as a result of national

cultural differences provide novel insights. Similarly, differences in attitudes towards tracking

as a result of cultural as well as structural differences shed further light on the influence of

national culture and industry structure on customer perceptions of loyalty programme

attributes. Despite the extensive body of existing research on the influence of loyalty

programmes and their attributes on customer loyalty, the influence of both national culture

and industry structure remains unexplored. Therefore, these findings address this research gap

by providing the first focused insight into how these two factors influence customer

expectations of loyalty programmes attributes.

Fourthly, the application of Dick and Basu’s (1994) loyalty typologies provides a more detailed

account of how national culture and industry structure influence customer loyalty. Compared,

to the previous studies that use customer loyalty as a holistic construct, the application of

different loyalty typologies provides a greater understanding of the differential influence

exerted by various cultural and structural elements. For instance, greater True Loyalty in the

UK as a result of retailer focus on quality and Spurious Loyalty towards some retailers in Sri

Lanka as a result of a lack of emphasis on customer service and quality underline this. In so

doing, this study extends the limited extant knowledge in the field of cross-cultural and cross-

structural customer loyalty in grocery retailing. Furthermore, these findings provide a starting

point for future research to be conducted, applying these loyalty typologies.

From a methodological perspective, this study is the first to apply Dick and Basu’s (1994)

customer loyalty typologies in a grocery retail setting (Ngobo 2017). Previous studies have not

explored differences in customer loyalty to grocery retailers in the UK and Sri Lanka. Whilst no

known studies have explored customer loyalty to grocery retailers in Sri Lanka, previous

177

studies in the UK; Omar and Sawmong (2007) as well as Doherty and Nelson (2008) have only

focused on customer loyalty in general. The application of this model clearly highlights the

existence of different categories of customer loyalty and the following contributions can thus

be identified;

The identification of these categories provides a more in-depth understanding of the

influence of national culture and industry structure on customer loyalty. Findings

clearly show how these two elements influence these categories. This insight provides

a more detailed understanding compared to previous studies that have used

customer loyalty as a general construct

The identification of different customer loyalty types highlights that customers can be

segmented based on their type and degree of loyalty; allowing grocery retailers to

better identify and reward the most loyal customers. This also allows retailers to

monitor customers so that shifts in loyalty and their underlying reasons could be

understood as proposed by Baloglu (2002) and Ngobo (2017)

This study addresses previous scholarly recommendations calling for Dick and Basu’s

(1994) model to be applied in the context of grocery retail. In doing so, this study

confirms the suitability of this model in light of concerns regarding its validity due to

its purely conceptual nature (Ngobo 2017)

A further methodological contribution can be attributed to the questionnaire items used in

this study. 53 of the 63 questions used to measure the various constructs were developed

following the qualitative findings. The validity and reliability of these questionnaire items were

confirmed by the exploratory factor analyses and confirmatory factor analyses in both

countries. Therefore, these items could be confidently applied to new measurement

instruments when conducting future research.

Finally, the theoretical framework (Figure 6.6) is the first such framework illustrating the

influence of national culture as well as industry structure on customer loyalty. Whilst previous

studies do not provide detailed illustrative frameworks, this framework summarises the key

findings and highlights the key relationships. Further, this framework illustrates the various

linkages; highlighting the direct influence of national culture and industry structure as well as

how these two factors influence customer loyalty indirectly through other elements. As a

result, the theoretical framework provided in this study can be argued to have greater

explanatory power compared to the more simplistic models proposed in previous studies. As a

178

result, this framework could be more appropriate to be used in future research in this area and

also to be applied in other retail settings with suitable adjustments.

7.5 Practical Implications & Recommendations to Grocery Retailers

The key findings in this study can be used to provide clear and practical recommendations to

grocery retailers due to its comparative and applied nature. Recommendations can be used

when designing products and service offerings as well as loyalty programmes in their efforts to

build and enhance customer loyalty.

7.5.1 Differences in Loyalty

The eight main grocery retailers in the UK need to continue to offer high levels of product

quality to customers so that their attitudinal loyalty will remain relatively high; a prerequisite

for true loyalty. It is clear that Marks & Spencer and Waitrose enjoy a niche position in the

market. These two retailers may be able to sustain their position as the preferred retailers by

maintaining focus on premium quality and customer service; sustaining an overall premium

image. In doing so, they will be able to maintain their existing customer base and attract

latency from other customers. However, these two retailers need to be mindful of some of the

concerns identified in the primary data about the gulf between price and quality. Therefore,

they need to ensure value for money so that the higher prices charged are justified by greater

quality. The Big Four and the two discounters may be content with high levels of true loyalty

reported towards them but need to ensure that they continue to offer high levels of product

quality to customers. Whilst attempting to emulate the premium retailers may not be

effective, they can build on the value for money proposition which is crucial to UK customers.

For domestic as well as international retailers focusing on Sri Lanka; it is likely that premium

grocery retailers may be well received by customers due to positive perceptions towards

retailers such as Keells and Arpico. In the absence of discounters in the country, there could be

potential for retailers such as Aldi and Lidl to cater to different segments of the market.

However, the existing customer base may not perceive such retailers positively due to low

perceptions towards the market leader; Cargills which takes an EDLP approach. With organised

retailing still in its infancy in rural areas of the country, customers in these areas with lower

incomes could in fact be more attracted to such discounters. Most importantly, grocery

retailers need to focus on quality as well as customer service if they are to enjoy true loyalty in

the country. These findings could be used by grocery retailers planning entry into other South

Asian countries that share cultural and structural similarities with Sri Lanka. For instance,

179

customer service may be equally valued by customers in neighbouring countries such as India,

Pakistan and Bangladesh due to their high Power Distance culture (Hofstede Insights 2018).

7.5.2 Loyalty Programmes

Grocery retailers can use loyalty programmes to track customers in Sri Lanka. There is

potential to gather data and target customers with individualised offers based on their

purchasing behaviour. Whilst this is not done at present by domestic grocery retailers,

international retailers such as Tesco and Sainsbury’s will be able to effectively use their

existing expertise and systems to implement such initiatives. Tracking could therefore provide

the opportunity to develop relationships with customers via two-way communication (Meyer-

Waarden and Benavent 2009). On the contrary, such tracking may not be as suitable in the UK.

In such instances, grocery retailers can use privacy-aware programmes where customers get to

decide how much personal information is disclosed to retailers (Blanco-Justicia and Domingo-

Ferrer 2016). Retailers then need to offer additional rewards to obtain more information from

customers. To mitigate privacy concerns, grocery retailers need to ensure customers that their

privacy will not be compromised by participating in a loyalty programme (Enzmann and

Schneider 2005). Furthermore, greater transparency should be practiced by disclosing to

customers how their data will be used. Ensuring high levels of customer perceived value will

also help grocery retailers in such contexts due to its positive impact on customer willingness

to disclose information (Leppäniemi et al. 2017). This is because customers weigh the risks and

benefits of participating in loyalty programmes (Norberg et al. 2007; John et al. 2011).

Therefore, the benefits of sharing their information should be clearly communicated to

customers (Taylor et al. 2015).

Instant or segregated rewards would be more effective in both countries. However, grocery

retailers need to evaluate how such strategies fit in with their overall objectives due to

concerns about these types of rewards. Whilst these may be effective in the short-term, they

may not serve to reduce incidence of customer switching. Finally, there is greater potential for

soft benefits, preferential treatment and tiered schemes in Sri Lanka. However, care should be

taken to avoid the possible negative reactions from customers who do not receive such

benefits and preferential treatment. It should also be noted that some customers may not

appreciate preferential treatment and such initiatives could cause embarrassment and

displeasure. Similarly, tiered schemes should be implemented with caution; mainly due to the

possible negative effect of customer demotions. On the contrary, hard or economic benefits

would be more suitable in the UK. Overall, grocery retailers need to understand and offer the

types of rewards that are perceived favourably by customers (Omar et al. 2010). Offering the

180

right rewards to customers will result in customer satisfaction and loyalty (Berenguer-Contri et

al. 2009). Therefore, focused research would be useful to identify individual customer

preferences in both countries.

7.5.3 Own Labels

Although moderate aversion to own labels in the UK appears to be a result of their tiered

nature, grocery retailers need to maintain this structure. This will allow them to cater to price

driven customers as well as those seeking more premium ranges. In light of recent research

highlighting customer preference for premium own labels (Mintel 2017), retailers need to

focus more on such products. There is an opportunity for grocery retailers to successfully

expand their own label ranges in Sri Lanka. Given their openness to such products, Sri Lankan

customers may be receptive to international grocery retailers with established own label

ranges. There is also potential to create a first mover advantage by introducing tiered labels in

the country. However, retailers need to be mindful of the possible perceptual differences in

relation to tiered products. Grocery retailers should strive to provide adequate quality to

customers, regardless of the tiers being offered. This is important as positive quality

perceptions have been found to result in loyalty to own labels as well as the retailer (Nies and

Natter 2012; Beneke et al. 2015; Coelho de Vale et al. 2016).

Whilst own labels should remain a priority for both domestic and international grocery

retailers, the importance of national brands should not be ignored. The provision of national

brands will ensure that an adequate assortment is offered to customers; especially those who

prefer such brands. this is crucial, particularly in light of some customer latency towards

national brands in the UK. Failing to do so could lead to customer dissatisfaction and perhaps

switching. Whilst own labels will ensure higher profit margins and market penetration;

Koschate-Fischer et al. (2014), national brands will bring in greater revenues and attract

customers who prefer such brands; Pepe et al. (2011). Finally, grocery retailers need to ensure

that their own labels are trusted by customers. Trust in own labels has been found to enhance

a retailer’s image and reputation; thereby leading to greater perceived quality and customer

loyalty (Calvo Porral and Lang 2015). Negative media coverage of the 2013 horsemeat scandal

(BBC 2013) and the more recent Fipronil egg scare (BBC 2017) for instance, highlights the

importance of customer trust.

181

7.5.4 Convenience Stores

Although convenience related factors are highly important in both countries, there is greater

potential in Sri Lanka mainly due to the underdeveloped nature of the convenience sector in

the country. Grocery retailers have the opportunity to expand this format and target more

residential areas in urban and suburban parts of the country. Such developments could be

targeted at top-up shopping which is currently being done in smaller local stores. Due to space

limitations in urban and suburban areas of the country; AT Kearney (2016) suggests that

targeting residential areas could be a viable option to penetrate the market and capture

market share from the unorganised grocery sector. Incoming grocery retailers with established

convenience formats have an opportunity to exploit this growing need by using their

experience to develop this sector.

The discussion of these aspects provides crucial guidance to domestic and international

grocery retailers. It can be noted that a universal application of product and service offerings

or loyalty programme attributes would not be appropriate. Consequently, grocery retailers;

both domestic and international, need to consider differences in cultural and structural

differences when entering new markets and when dealing with increasingly diverse domestic

Western markets.

7.6 Research Scope & Limitations

The mixed methods approach taken in this study ensured that the qualitative findings would

provide a preliminary insight into the influence of national culture and industry structure. In

the absence of adequate previous knowledge, these findings provided the base for the ensuing

quantitative phase. As a result, the development of the questionnaire was guided by the key

qualitative findings. The language and wording used in the questionnaire were grounded in the

qualitative findings; ensuring complete clarity and respondent comprehension. Furthermore,

the qualitative data aided the interpretation and discussion of the quantitative findings by

providing additional insights into certain phenomena. This process of ‘abduction’ (Johnson and

Onwuegbuzie 2004) enabled the researcher to go back and forth between the two sets of data

to ensure a more in-depth picture of the phenomena under study. This also enabled a richer

understanding of the data as the qualitative findings in particular, complemented the

statistical results.

The scope of this study meant that some of the insights could not be substantiated through

additional primary data. For instance, the moderate aversion to own labels in the UK was

argued to be due to the tiered structure of these products. Although such inference was

182

guided by secondary data, more focused primary data would have provided greater predictive

validity to the interpretation of this finding. In a similar manner, some of the posited

relationships; the mediating influence of national culture on customer loyalty via loyalty

programmes for instance was done so based on secondary data. Likewise, aspects such as

aversion to own labels and fresh produce were also not tested against the respondents’

respective main store. The general nature of these constructs negated the ability to conduct

conclusive statistical tests such as regressions to verify their influence on customer loyalty.

Although the UK survey sample was drawn from the Postcode Finder; ensuring adequate

representation of the population, a different mode of administration may have yielded a

higher response rate. Postal questionnaires resulted in a response rate of 10% and whilst

tactics such as naming, incentives and reminders may have improved the rate of response, the

efficacy of this method should be questioned. Given the high internet penetration rate in the

UK, email surveys may be a suitable alternative although self-selection bias is an inherent

limitation of such methods (Bethlehem 2010). Support for digital administration methods is

evident from the satisfactory response rates to the online survey used in Sri Lanka; despite the

relatively low levels of internet penetration in the country. A random sample was not

obtainable in Sri Lanka due to the unavailability of a sampling frame. Although every effort was

made by the researcher to obtain as a broad a sample as possible, it should be noted that the

purposive snowball sampling technique used may have introduced some bias to the sample.

7.7 Recommendations for Future Research

The theoretical framework provided in this study allows a detailed understanding of the

influence of national culture and industry structure on customer loyalty in grocery retailing.

This framework is relevant in the much broader retail context and holds potential to be applied

in other areas of retail consumer behaviour such as fashion. Therefore, future research could

be conducted in grocery retailing as well as other retail contexts to test and validate this

framework. Future research should also focus on adding more elements to this model to

improve its overall explanatory power.

However, future research needs to obtain adequate primary data to cover the various

relationships identified in this framework. This will enable the relationships to be tested using

techniques such as regression; allowing the further investigation of the effect of the various

causal factors identified. Doing so will enhance the predictive validity as well as the

explanatory power of this framework; making it more universally applicable in retail consumer

behaviour research. Similarly, follow-up qualitative studies with the respondents in this study

183

could allow a greater understanding of these findings as well as the various relationships and

their likely causes. For instance, the role of tiered own labels in the UK could be explored in

greater detail. In a similar manner, aversion to or openness to various loyalty programme

attributes could be explained through more focused qualitative research.

This study only captured two countries and whilst these findings are credible, future studies

could involve more countries with cultural and structural differences. Doing so would allow a

more holistic and comprehensive argument to be developed. This framework could be tested

in other South Asian countries to ascertain its applicability across the region. Similarly, its

application to other Western countries with similar national cultural dimensions and industry

structural development levels would help establish its generalisability. Thereafter, responses

could be obtained from customers in each of these countries; enabling the development and

testing of a statistical model.

184

Reference List

AT Kearney, 2016. The 2016 Global Retail Development Index. Global Retail Expansion at a

Crossroads [online]. Chicago: AT Kearney. Available from:

https://www.atkearney.com/documents/10192/8226719/Global+Retail+Expansion+at+a+Cros

sroads%E2%80%932016+GRDI.pdf/dc845ffc-fe28-4623-bdd4-b36f3a443787 [Accessed 10

January 2018].

AT Kearney, 2017. The Age of Focus [online]. Chicago: AT Kearney. Available from:

https://www.atkearney.com/global-retail-development-index/rankings [Accessed 14 January

2018].

Abubakar, M.M., Mokhtar, S.S.M. and Abdullateef, A.O., 2013. The Moderating Effect of Long-

Term Orientation Culture on the Relationship Between Trust, Personalization and Customer

Satisfaction and Loyalty: A Proposed Framework. International Journal of Academic Research in

Business and Social Sciences [online], 3 (9), 117-131.

Ackerman, D. and Tellis, G., 2001. Can Culture Affect Prices? A Cross-Cultural Study of

Shopping and Retail Prices. Journal of Retailing [online], 77, 57-82.

Acocella, I., 2012. The focus Groups in Social Research: Advantages and Disadvantages. Quality

& Quantity [online], 46 (4), 1125-1136.

Adelson, J.L. and McCoach, D.B., 2010. Measuring the Mathematical Attitudes of Elementary

Students: The Effects of a 4-Point or 5-Point Likert-Type Scale. Educational and Psychological

Measurement [online], 70 (5), 796-807.

Ahluwalia, R.H., Unnava, H.R. and Burnkrnat, R.E., 1999. Understanding the Value of a Loyal

Customer: An Information-Processing Perspective. Cambridge, Massachusetts: Marketing

Science Institute. Report 37-38.

Akehurst, G. and Alexander, N., 1995. Developing a Framework for the Study of the

Internationalisation of Retailing. Service Industries Journal [online], 15 (4), 204-209.

Albala, K., 2011. Food cultures of the world encyclopedia, Santa Barbara: ABC-CLIO, LLC.,

Alexander, N., 1990. Retailers and International Markets: Motives for Expansion. International

Marketing Review [online], 7 (4), 75-85.

185

Alexander, N. and Doherty, A.M., 2009. International Retailing. Oxford: Oxford University

Press.

Allen, I.E. and Seaman, C.A., 2007. Liker Scales and Data Analyses. Quality Progress [online], 40

(7), 64-65.

Amine, A., 1998. Consumer’s True Brand Loyalty: The Central Role of Commitment. Journal of

Strategic Marketing [online], 6 (4), 305-319.

Anderson, J.C. and Gerbing, D.W., 1988. Structural equation modeling in practice: a review and

recommended two-step approach. Psychological Bulletin [online], 103 (3), 411-423.

Arbore, A. and Estes, Z., 2013. Loyalty program structure and consumers’ perceptions of

status: Feeling special in a grocery store? Journal of Retailing and Consumer Services [online],

20 (5), 439-444.

Arnold, K.A. and Bianchi, C., 2001. Relationship Marketing, Gender, and Culture: Implications

for Consumer Behavior. Advances in Consumer Research [online], 28, 100-105.

Arpico, 2018. Expertise - Retail [online]. Nawinna: Richard Peiris & Company PLC. Available

from: https://www.arpico.com/contents/our_expertise_retail.php#.WldaVldSnTc [Accessed 10

January 2018].

Assaf, A.G., Josiassen, A., Ratchford, B.T. and Barros, C.P. 2012. Internationalization and

Performance of Retail Firms: A Bayesian Dynamic Model. Journal of Retailing [online], 88 (2),

191-205.

Athanasopoulou, P., 2009. Relationship Quality: A Critical Literature Review and Research

Agenda. European Journal of Marketing [online], 43 (5/6), 583-610.

Bagozzi. R.P., 1995. Reflections on Relationship Marketing in Consumer Markets. Journal of the

Academy of Marketing Science [online], 23 (4), 272-77.

Bagozzi, R.P. and Yi, Y., 1988. On the evaluation of structural equation models. Journal of the

Academy of Marketing Science [online], 16 (1), 74-94.

Bahhouth, V., Ziemnowicz, C. and Zgheib, Y., 2012. Effect of Culture and Traditions on

Consumer Behavior in Kuwait. International Journal of Business, Marketing, & Decision Science

[online], 5 (2), 1-11.

186

Baldinger, A.L. and Rubinson, J., 1996. Brand Loyalty: The Link Between Attitude and Behavior.

Journal of Advertising Research [online], 36 (6), 22-34.

Baloglu, S., 2002. Dimensions of Customer Loyalty: Separating Friends from Well Wishers.

Cornell Hotel and Restaurant Administration Quarterly [online], 43, 47-59.

Baltas, G., Argouslidis, P.C. and Skarmeas, D., 2010. The Role of Customer Factors in Multiple

Store Patronage: A Cost–Benefit Approach. Journal of Retailing [online], 86, 37-50.

Bandyopadhyay, S. and Martell, M., 2007. Does Attitudinal Loyalty Influence Behavioral

Loyalty? A Theoretical and Empirical Study. Journal of Retailing & Consumer Services [online],

14, 35-44.

Bao, Y., Sheng, S., Bao, Y. and Stewart, D., 2011. Assessing quality perception of private labels:

intransient cues and consumer characteristics. Journal of Consumer Marketing [online], 28 (6),

448-458.

Barclay, S., Todd, C., Finlay, I., Grande, G. and Wyatt, P., 2002. Not another questionnaire!

Maximizing the response rate, predicting non-response and assessing non-response bias in

postal questionnaire studies of GPs. Family Practice [online], 19 (1), 105-111.

BBC, 2010. Safeway scraps loyalty card [online]. London: British Broadcasting Corporation.

Available from: http://news.bbc.co.uk/1/hi/business/735835.stm [Accessed 10 January 2018].

BBC, 2013. Horsemeat scandal: Tesco reveals 60% content in dish [online]. London: British

Broadcasting Corporation. Available from: http://www.bbc.co.uk/news/uk-21418342

[Accessed 14 January 2018].

BBC, 2017. Fipronil egg scandal: What we know [online]. London: British Broadcasting

Corporation. Available from: http://www.bbc.co.uk/news/world-europe-40878381 [Accessed

14 January 2018].

Bearden, W.O., 1977. Determinant Attributes of Store Patronage: Downtown versus Outlying

Shopping Centers. Journal of Retailing [online], 53 (2), 15-22.

Beaton, D.E., Bombardier, C., Guillemin, F. and Ferraz, M.B., 2000. Guidelines for the process of

cross-cultural adaptation of self-report measures. Spine [online], 25 (24), 3186-3191.

187

Beck, J.T., Chapman, K. and Palmatier, R.W., 2015. Understanding Relationship Marketing and

Loyalty Program Effectiveness in Global Markets. Journal of International Marketing [online],

23 (3), 1-21.

Beneke, J., Brito, A. and Garvey, K.A., 2015. Propensity to buy private label merchandise The

contributory effects of store image, price, risk, quality and value in the cognitive stream.

International Journal of Retail & Distribution Management [online], 43 (1), 43-62.

Berenguer-Contri, G., Ruiz-Molina, M.E. and Gil-Saura, I., 2009. Relationship Benefits and Costs

in Retailing: A Cross-Industry Comparison. Journal of Retail and Leisure Property [online], 8, 57-

66.

Berry, L.L., 1983. Relationship Marketing [online]. In: Berry, L.L., Shostack, G.L. and Upah, G.D.,

eds. Emerging Perspectives of Services Marketing. American Marketing Association, Chicago,

IL, 1983. 25-38.

Berry, L.L., 1995. Relationship Marketing of Services: Growing Interest, Emerging Perspectives.

Journal of the Academy of Marketing Science, 23 (4), 236-245.

Berry, L.L., 2002. Relationship Marketing of Services- Perspectives from 1983 and 2000. Journal

of Relationship Marketing [online], 1, 59-77.

Berzelak, N., Vehovar, V. and Lozar Manfreda, K., 2015. Web Mode as Part of Mixed-Mode

Surveys of the General Population: An Approach to the Evaluation of Costs and Errors.

Advances in Methodology & Statistics [online], 12 (2), 45-68.

Bethlehem, J., 2010. Selection Bias in Web Surveys. International Statistical Review [online], 78

(2), 161-188.

Biesta, G., 2010. Pragmatism and the Philosophical Foundations of Mixed Methods Research.

In: Tashakkori, A. and Teddlie, C., eds. SAGE Handbook of Mixed Methods in Social and

Behavioral Research. 2nd edition. Thousand Oaks: SAGE Publications, Inc., 95-118.

Blanco-Justicia, A. and Domingo-Ferrer, J., 2016. Privacy-aware loyalty programs. Computer

Communications [online], 82 (2016), 83-94.

Bloemer, J. and de Ruyter, K., 1998. On the Relationship Between Store Image, Store

Satisfaction and Store Loyalty. European Journal of Marketing [online], 32 (5/6), 499-513.

188

Blut, M., Evanschitzky, H., Vogel, V. and Ahlert, D., 2007. Switching Barriers in the Four-Stage

Loyalty Model. Advances in Consumer Research [online], 34, 726-734.

Blut, M., Teller, C. and Floh, A., 2018. Testing Retail Marketing-Mix Effects on Patronage: A

Meta-Analysis. Journal of Retailing [online], In Press.

Boedeker, M., 1997. Relationship Marketing and Regular Customer Cards: Daily Product

Retailing in Finland. Marketing Intelligence & Planning [online], 15 (6), 249-257.

Boukhobza, T., 2005. An Investigation into Customer Loyalty and Relationship Marketing: Case

UK Grocery Retailing and Telecommunication Sectors [online]. Thesis (PhD). University of

Salford.

Bove, L.L. and Johnson, L.W., 2009. Does "true" personal or service loyalty last? A longitudinal

study. Journal of Services Marketing [online], 23 (3), 187-194.

Boyatzis, R.E., 1998. Transforming Qualitative Information: Thematic Analysis and Code

Development. Thousand Oaks: SAGE Publications, Inc.

Braun, V. and Clarke, V., 2006. Using Thematic Analysis in Psychology. Qualitative Research in

Psychology [online], 3 (2), 77-101.

Bray, J.P., 2011. A Critical Evaluation into the Role of Ethics in Clothing Purchase Decisions

[online]. Thesis (PhD). Bournemouth University.

Bridson, K., Evans, J. and Hickman, M., 2008. Assessing the relationship between loyalty

program attributes, store satisfaction and store loyalty. Journal of Retailing and Consumer

Services [online], 15 (5), 364-374.

Brody, R.P. and Cunningham S.M., 1968. Personality Variables and the Consumer Decision

Process. Journal of Marketing Research [online], 5, 50-57.

Broyles, S.A., 2009. Loyalty's Influence on Satisfaction in Cross-Cultural Settings. Journal of

Product & Brand Management [online], 18 (6), 414-424.

Bryman, A., 2007. Barriers to Integrating Quantitative and Qualitative Research. Journal of

Mixed Methods Research [online], 1, 8-22.

Bryman, A. and Bell, E., 2011. Business Research Methods. 3rd edition. Oxford: Oxford

University Press.

189

Budhathoki, T., 2014. The impact of culture on own-label brands performance [online]. Thesis

(PhD). Loughborough University.

Burt, S.L. and Sparks, L., 2003. Power and Competition in the UK Retail Grocery Market. British

Journal of Management [online], 14 (3), 237-254.

Burt, S.L., Davies, K., Dawson, J. and Sparks, L., 2008. Categorizing patterns and processes in

retail grocery internationalistion. Journal of Retailing and Consumer Services [online], 15 (2),

78-92.

Burton, S., Lichtenstein, D.R., Netemeyer, R.G. and Garretson, J.A., 1998. A Scale for Measuring

Attitude Toward Private Label Products and an Examination of Its Psychological and Behavioral

Correlates. Journal of the Academy of Marketing Science [online], 26 (4), 293-306.

Butori, R. and De Bruyn, A., 2013. So you want to delight your customers: The perils of ignoring

heterogeneity in customer evaluations of discretionary preferential treatments. International

Journal of Research in Marketing [online], 30 (4), 358-367.

Calvo Porral, C. and Lang, M.F., 2015. Private labels. The role of manufacturer identification,

brand loyalty and image on purchase intention. British Food Journal [online], 117 (2), 506-522.

Cargills (Ceylon) PLC, 2018. Cargills Store Locator [online]. Colombo: Cargills (Ceylon) PLC.

Available from: http://www.cargillsceylon.com/StoreLocator.aspx [Accessed 11 January 2018].

Cedrola, E. and Memmo, S., 2010. Loyalty Marketing and Loyalty Cards: A Study of the Italian

Market. International Journal of Retail and Distribution Management [online], 38 (3), 205-225.

Chan, P., Finnegan, C. and Sternquist, B., 2011. Country and firm level factors in

international retail expansion. European Journal of Marketing [online], 45 (6), 1005-1022.

Chang, L., 1994. A Psychometric Evaluation of 4-Point and 6-Point Likert-Type Scales in Relation

to Reliability and Validity. Applied Psychological Measurement [online], 18 (3), 205-215.

Chang, C.H. and Tu, C.Y., 2005. Exploring Store Image, Customer Satisfaction and Customer

Loyalty Relationship: Evidence from Taiwanese Hypermarket Industry. Journal of American

Academy of Business [online], 7 (2), 197-202.

Chang, S., van Witteloostuijn, A. and Eden, L., 2010. From the Editors: Common Method

Variance in International Business Research. Journal of International Business Studies [online],

41 (2), 178-184.

190

Chen, F.F., 2007. Sensitivity of Goodness of Fit Indexes to Lack of Measurement Invariance.

Structural Equation Modeling [online], 14 (3), 464-504.

Cheung, G.W. and Rensvold, R.B., 2002. Evaluating Goodness-of-Fit Indexes for Testing

Measurement Invariance. Structural Equation Modeling [online], 9 (2), 233-255.

Christensen, A.I., Ekholm, O., Kristensen, P.L., Larsen, F.B., Vinding, A.L., Glumer, C. and Juel, K.,

2015. The effect of multiple reminders on response patterns in a Danish health survey.

European Journal of Public Health [online], 25 (1), 156-161.

Christopher, M., Ballantyne, D. and Payne, A., 1991. Relationship Marketing: Bringing Quality,

Customer Service and Marketing Together. Oxford: Butterworth-Heinemann.

Clark, T., 1990. International Marketing and National Character: A Review and Proposal for An

Integrative Theory. Journal of Marketing [online], 54 (4), 66-79.

Coelho de Vale, R., Verga Matos, P. and Caiado, J., 2016. The impact of private labels on

consumer store loyalty: An integrative perspective. Journal of Retailing and Consumer Services

[online], 28 (2016), 179-188.

Cohen, J., 1988. Statistical Power Analysis for the Behavioral Sciences. 2nd edition. New Jersey:

Lawrence Erlbaum Associates.

Colliander, J., Söderland, M. and Szugalski, S., 2016. Multi-level loyalty program rewards and

their effects on top-tier customers and second-tier customers. Journal of Consumer Marketing

[online], 33 (3), 162-171.

Collins English Dictionary, 2018. Culture [online]. London: Harper Collins Publishers. Available

from: http://www.collinsdictionary.com/dictionary/english/culture?showCookiePolicy=true


Copeland, M.T., 1923. Relation of Consumers' Buying Habits to Marketing Methods. Harvard

Business Review [online], 1 (3), 282-289.

Cottrell, E., Roddy, E., Rathod, T., Thomas, E., Porcheret, M. and Foster, N.E., 2015. Maximising

response from GPs to questionnaire surveys: do length or incentives make a difference? BMC

Medical Research Methodology [online], 15, 39-50.

Creswell, J.W. and Plano Clark, V.L., 2011. Designing and Conducting Mixed Methods Research.

2nd edition. Los Angeles: SAGE Publications, Inc.

191

Creswell, J.W., 2013. Research Design: Qualitative, Quantitative, And Mixed Method

Approaches. 3rd edition. Los Angeles: SAGE Publications, Inc.

Croasmun, J.T. and Ostrom, L., 2011. Using Likert-Type Scales in the Social Sciences. Journal of

Adult Education [online], 40 (1), 19-22.

Cronin, J.J. and Taylor, S.A., 1992. Measuring Service Quality: A Re-Examination and Extension.

Journal of Marketing [online], 56 (3), 55-68.

Crosby, L.A., Evans, K.R. and Cowles, D., 1990. Relationship Quality in Services Selling: An

Interpersonal Influence Perspective. Journal of Marketing [online], 54 (3), 68-81.

Crowley, G.H. and Gilreath, C. L., 2002. Probing User Perceptions of Service Quality: Using

Focus Groups to Enhance Quantitative Surveys. Performance Measurement & Metrics [online],

3 (2), 78-84.

Cunningham, R.M., 1961. Customer Loyalty to Store and Brand. Harvard Business Review

[online], 39 (6), 127-137.

Daily News, 2017. HUGO BOSS, launched in Sri Lanka [online]. Colombo: Lake House. Available

from: http://dailynews.lk/2017/05/05/business/115032/hugo-boss-launched-sri-lanka


Das, G., 2014. Impact of Store Image on Store Loyalty and Purchase Intention: Does it vary

across Gender? International Journal of Electronic Marketing and Retailing [online], 6, 52-71.

Dash, S., Bruning, E. and Acharya, M. 2009. The effect of power distance and individualism on

service quality expectations in banking. A two-country individual- and national-cultural

comparison. International Journal of Bank Marketing [online], 27 (5), 336-358.

Dawkins, P.M. and Reichheld, F.F., 1990. Customer Retention as a Competitive Weapon.

Directors & Boards [online], 14 (4), 42-47.

Day, G.S., 1969. A Two-Dimensional Concept of Brand Loyalty. Journal of advertising Research

[online], 9 (3), 29-35.

DelVecchio, D., 2001. Consumer perceptions of private label quality: the role of product

category characteristics and consumer use of heuristics. Journal of retailing and Consumer


192

de Mooij, M. and Hofstede, G., 2002. Convergence and Divergence in Consumer Behavior:

Implications for International Retailing. Journal of Retailing [online], 78, 61-69.

de Mooij, M., 2004. Consumer Behavior and Culture: Consequences for Global Marketing and

Advertising. London: SAGE Publications, Inc.

de Mooij, M., 2010. Global Marketing and Advertising: Understanding Cultural Paradoxes.

London: SAGE Publications, Inc.

de Mooij, M. and Hofstede, G., 2010. The Hofstede Model. International Journal of Advertising

[online], 29 (1), 85-110.

de Mooij, M. and Hofstede, G., 2011. Cross-Cultural Consumer Behavior: A Review of Research

Findings. Journal of International Consumer Marketing [online], 23 (3/4), 181-192.

de Mooij, M., 2017.Comparing dimensions of national culture for secondary analysis of

consumer behavior data of different countries. International Marketing Review [online], 34

(3), 444-456.

Demoulin, N.T.M. and Zidda, P., 2009. Drivers of Customers’ Adoption and Adoption Timing of

a New Loyalty Card in the Grocery Retail Market. Journal of Retailing [online], 85 (3), 391-405.

Denscombe, M., 2008. Communities of Practice: A Research Paradigm for the Mixed Methods

Approach. Journal of Mixed Methods Research [online], 2 (3), 270-283.

Department of Census and Statistics - Sri Lanka, 2017. Household Income and Expenditure

Survey – 2016 [online]. Battaramulla: Department of Census and Statistics. Available from:

http://www.statistics.gov.lk/HIES/HIES2016/HIES_2016_final_Result.pdf [Accessed 12 January

2018].

Department of Census and Statistics - Sri Lanka, 2018a. Mid-year Population Estimates by

District & Sex, 2012 - 2017 [online]. Battaramulla: Department of Census and Statistics.

Available from:

http://www.statistics.gov.lk/PopHouSat/VitalStatistics/MidYearPopulation/Mid-

year%20population%20by%20district.pdf [Accessed 12 January 2018].

Department of Census and Statistics - Sri Lanka, 2018b. Mid-year Population Estimates by Age

Group and Sex, 2012 - 2017 [online]. Battaramulla: Department of Census and Statistics.

Available from:

193

http://www.statistics.gov.lk/PopHouSat/VitalStatistics/MidYearPopulation/Mid-

year%20population%20by%20age%20group.pdf [Accessed 12 January 2018].

De Wulf, K., Odekerken-Schroder, G. and Iacobucci, D., 2001. Investments in Consumer

Relationships: A Cross-Country and Cross-Industry Exploration. Journal of Marketing [online],

65 (4), 33-50.

De Wulf, K. and Odekerken-Schroder, G., 2003. Assessing the Impact of a Retailer’s

Relationship Efforts on Consumers’ Attitudes and Behavior. Journal of Retailing and Consumer


Dick, A.S. and Basu, K., 1994. Customer Loyalty: Toward an Integrated Conceptual Framework.

Journal of the Academy of Marketing Science [online], 22 (2), 99-113.

Dillman, D.A., Lesser, V., Mason, R., Carlson, J., Willits, F., Robertson, R. and Burke, B., 2007.

Personalization of Mail Surveys for General Public and Populations with a Group Identity:

Results from Nine Studies. Rural Sociology [online], 72 (4), 632-646.

Dillman, D.A., Smyth, J.D. and Christian, L.M., 2014. Internet, Phone, Mail, and Mixed-Mode

Surveys : The Tailored Design Method. 4th edition. New Jersey: Wiley.

Doherty, S. and Nelson, R., 2008. Customer loyalty to food retailers in Northern Ireland:

‘devoted loyals’ or ‘promiscuous switchers’? International Journal of Consumer Studies

[online], 32 (2008), 349-355.

Doney, P.M., Cannon, J.P. and Mullen, M.R., 1998. Understanding the Influence of National

Culture on the Development of Trust. Academy of Management Review [online], 23 (3), 601-

620.

Donthu, N. and Yoo, B., 1998. Cultural Influences on Service Quality Expectations. Journal of

Service Research [online], 1 (2), 178-186.

Dowling, G.R. and Uncles, M., 1997. Do Customer Loyalty Programs Really Work? Sloan

Management Review [online], 38 (4), 71-82.

Doyle, P. and Fenwick, I., 1974. How Store Image Affects Store Shopping Habits in Grocery

Chains. Journal of Retailing [online], 50 (4), 39-52.

Doyle, L., Brady, A. and Byrne, G., 2009. An Overview of Mixed Methods Research. Journal of

Research in Nursing [online], 14 (2), 175-185.

194

Dreze, X. and Nunes, J.C., 2009. Feeling Superior: The Impact of Loyalty Program Structure on

Consumers’ Perceptions of Status. Journal of Consumer Research [online], 35 (6), 890-905.

Dubois, B. and Laurent, G., 1999. A Situational Approach to Brand Loyalty. In: Arnould, E.J. and

Scott, L.M., eds. Advances in Consumer Research. Association for Consumer Research, Provo,

Utah, 1999. 657-663.

Duffy, D.L., 2003. Commentary: Internal and External Factors which Affect Customer Loyalty.

Journal of Consumer Marketing [online], 20 (5), 480-485.

Duke, R.C., 1989. A Structural Analysis of the UK Grocery Retail Market. British Food Journal

[online], 91 (5), 17 22.

Duke, R., 1992. Structural Changes in the UK Grocery Retail Market. British Food Journal

[online], 94 (2), 18-23.

Dunn, R. and Wrigley, L., 1985. Store Loyalty for Grocery Products: An Empirical Study. Area

[online], 16 (4), 307-314.

East, R., Harris, P., Willson, G. and Lomax, W., 1995. Loyalty to Supermarkets. International

Review of Retail Distribution & Consumer Research [online], 5, 99-109.

East, R., Harris, P., Lomax, W., Willson, G. and Hammond, K., 1998. Customer Defection From

Supermarkets. Advances in Consumer Research [online], 25, 507-512.

Eckman, M., Sakarya, S., Hyllegard, K., Gomez Borja, M.A. and Descals, A.M., 2015. Consumer

receptiveness to international retail expansion: a cross-cultural study of perceptions of social

and economic influence of foreign retailers. The International Review of Retail, Distribution and

Consumer Research [online], 25 (3), 260-275.

Ehrenberg, A.S.C., 1988. Repeat Buying Facts, Theory and Applications. New York: Oxford

University Press.

El-Manstrly, D., Paton, R., Veloutsou, C. and Moutinho, L., 2011. An empirical investigation of

the relative effect of trust and switching costs on service loyalty in the UK retail banking

industry. Journal of Financial Services Management [online], 16 (2), 101-110.

Enis, B.M. and Paul, G.W., 1970. 'Store Loyalty' as a Basis for Market Segmentation. Journal of

Retailing [online], 46 (3), 42-56.

195

Enzmann, M. and Schneider, M., 2005. Improving Customer Retention in E-Commerce through

a Secure and Privacy-Enhanced Loyalty System. Information Systems Frontiers [online], 7 (4-5),

359-370.

Etgar, M. and Rachman-Moore, D., 2008. International expansion and retail sales: an empirical

study. International Journal of Retail & Distribution Management [online], 36 (4), 241-259.

Euromonitor International, 2014. Markets of the Future in Sri Lanka [online]. London:

Euromonitor International. Available from: http://www.euromonitor.com/markets-of-the-

future-in-sri-lanka/report [Accessed 11 January 2018].

Evanschitzky, H., Ramaseshan, B., Woisetschläger, D., Richelsen, V., Blut, M. and Backhaus, C.,

2012. Consequences of Customer Loyalty to the Loyalty Program and to the Company. Journal

of the Academy of Marketing Science [online], 40 (5), 625-638.

Farley, J.U., 1964. Why Does Brand Loyalty Vary Over Products. Journal of Marketing Research

[online], 1, 9-14.

Feilzer, M.Y., 2010. Doing Mixed Methods Research Pragmatically: Implications for the

Rediscovery of Pragmatism as a Research Paradigm. Journal of Mixed Methods Research

[online], 4 (1), 6-16.

Fern, E.F., 2001. Advanced Focus Group Research. 1st edition. London: SAGE Publications, Inc.

Field, A.P., 2013. Discovering statistics using IBM SPSS statistics: and sex and drugs and rock 'n'

roll. 4th edition. Los Angeles: SAGE Publications Inc.

Filipe, S., Marques, S.H. and Salgueiro, M.d.F., 2017. Customers’ relationship with their grocery

store: Direct and moderating effects from store format and loyalty programs. Journal of

Retailing and Consumer Services [online], 37 (1), 78-88.

Financial Times, 2014. International designer brand Tommy Hilfiger to launch in Sri Lanka

[online]. Colombo: Wijeya Newspapers Ltd. Available from:

http://www.ft.lk/fashionlifestyle/international-designer-brand-tommy-hilfiger-to-launch-in-sri-

lanka/10404-369417 [Accessed 11 January 2018].

Financial Times, 2017. Why South Asia makes a good match for Britain after Brexit [online].

London: Financial Times Ltd. Available from: https://www.ft.com/content/999adb1c-e3b7-

11e6-8405-9e5580d6e5fb [Accessed 11 January 2018].

196

Financial Times, 2018. PM opens first Lanka Sathosa Mega store [online]. Colombo: Wijeya

Newspapers Ltd. Available from: http://www.ft.lk/front-page/PM-opens-first-Lanka-Sathosa-

Mega-store/44-646686 [Accessed 11 January 2018].

Food Standards Agency, 2016. The Food & You Survey Wave 4 [online]. London: Food

Standards Agency. Available from: https://www.food.gov.uk/sites/default/files/food-and-you-

w4-combined-report.pdf [Accessed 24 January 2018].

Fornell, C. and Larcker, D.F., 1981. Evaluating structural equation models with unobservable

variables and measurement error. Journal of Marketing Research [online], 18 (1), 39-50.

Gable, M., Fiorito, S.S. and Topol, M.T., 2008. An Empirical Analysis of the Components of

Retailer Customer Loyalty Programs. International Journal of Retail & Distribution

Management [online], 36, 32-49.

Ganguli, S. and Kumar, B.V., 2008. Drivers of Customer Satisfaction and Loyalty in Indian Retail

Supermarkets: An Exploratory Study. The Icfaian Journal of Management Research [online], 7

(12), 60-73.

Garbarino, E. and Johnson, M.S., 1999. The Different Roles of Satisfaction, Trust, and

Commitment in Customer Relationships. Journal of Marketing [online], 63 (2), 70-87.

Garcia-Gomez, B., Gutierrez-Arranz, A.M. and Gutierrez-Cillan, J., 2006. The Role of Loyalty

Programs in Behavioral and Affective Loyalty. Journal of Consumer Marketing [online], 23 (7),

387-396.

Garcia-Gomez, B., Gutierrez-Arranz, A.M. and Gutierrez-Cillan, J., 2012. Exploring the influence

of three types of grocery retailer loyalty programs on customer affective loyalty. The

International Review of Retail, Distribution and Consumer Research [online], 22 (5), 547-561.

Garland, R., 1991. The mid-point on a rating scale: Is it desirable? Marketing Bulletin [online],

2, 66-71.

Garson, D., 2015. Missing Values Analysis & Data Imputation. North Carolina: Statistical

Associates Publishing.

Gaskin, J., 2013. SEM Series Part 2: Data Screening. Available from:

https://www.youtube.com/watch?v=1KuM5e0aFgU [Accessed 12 January 2018].

197

Gaskin, J., 2017a. Data Screening. Available from:

http://statwiki.kolobkreations.com/index.php?title=Data_screening [Accessed 12 January

2018].

Gaskin, J., 2017b. Exploratory Factor Analysis. Available from:

http://statwiki.kolobkreations.com/index.php?title=Exploratory_Factor_Analysis [Accessed 12

January 2018].

Gaskin, J., 2017c. Confirmatory Factor Analysis. Available from:

http://statwiki.kolobkreations.com/index.php?title=Confirmatory_Factor_Analysis [Accessed

12 January 2018].

Geert Hofstede, 2018. Research and VSM [online]. Ede: Business Media. Available from:

http://geerthofstede.com/research-and-vsm/ [Accessed 24 January 2018].

Gendall, P., The effect of covering letter personalisation in mail surveys. International Journal

of Market Research [online], 47 (4), 367-382.

Ghosh, A., 1990. Retail Management. 2nd edition. Chicago, IL: The Dryden Press.

Goldman, A., 1974. Outreach of Consumers and the Modernization of Urban Food Retailing in

Developing Countries. Journal of Marketing [online], 38 (4), 8-16.

Goldman, A., 1977. The Shopping Style Explanation for Store Loyalty. Journal of Retailing

[online], 53 (4), 33-47.

Goldman, A., 1982. Adoption of Supermarket Shopping in a Developing Country: The Selective

Adoption Phenomenon. European Journal of Marketing [online], 16 (1), 17-26.

Google, 2018. Google Maps. Available from: http://maps.google.co.uk/ [Accessed 12 January

2018].

Greenbaum, T.L., 2000. Moderating Focus Groups: A Practical Guide for Group Facilitation. 1st

edition. Thousand Oaks: SAGE Publications, Inc.

Grewal, D., Baker, J., Levy, M. and Voss, G.B., 2003. The effects of wait expectations and store

atmosphere evaluations on patronage intentions in service-intensive retail stores. Journal of

Retailing [online], 79 (4), 259-268.

Grönroos, C. 1984. A Service Quality Model and its Marketing Implications. European Journal of

Marketing [online], 18 (4), 36-44.

198

Grönroos, C., 1995. Relationship Marketing: The Strategy Continuum. Journal of the Academy

of Marketing Science [online], 23 (4), 252-254.

Grosso, M. and Castaldo, S., 2015. How store attributes impact shoppers’ loyalty: do different

national cultures follow the same loyalty building process? The International Review of Retail,

Distribution and Consumer Research [online], 25 (5), 503-515.

Guba, E.G., 1990. The Alternative Paradigm Dialog. In: Guba, E.G., ed. The Paradigm Dialog.

Newbury Park: SAGE Publications, Inc., 17-27.

Guba, E.G. and Lincoln, Y.S., 2005. Paradigmatic Controversies, Contradictions, and Emerging

Confluences. In: Denzin, N.K. and Lincoln, Y.S., eds. The SAGE Handbook of Qualitative

Research. Thousand Oaks: SAGE Publications, Inc., 105-117.

Guenzi, P., Johnson, M.D. and Castaldo, S., 2009. A Comprehensive Model of Customer Trust in

Two Retail Stores. Journal of Service Management [online], 20 (3), 290-316.

Guest, G., 2013. Describing Mixed Methods Research: An Alternative to Typologies. Journal of

Mixed Methods Research [online], 7 (2), 141-151.

Guillemin, F., Bombardier, C. and Beaton, D., 1993. Cross-cultural adaptation of health-related

quality of life measures: Literature review and proposed guidelines. Journal of Clinical

Epidemiology [online], 46 (12), 1417-1432.

Guzi, M., Kahanec, M. and Kurekova, L.M. 2015. How Immigration Grease is Affected by

Economic, Institutional and Policy Contexts: Evidence from EU Labor Markets. Social Sciences

Research Network [online]. Available from: http://ftp.iza.org/dp9108.pdf [Accessed 14 January

2018].

Hair J.F., Black W.C., Babin J.B., Anderson R.E. and Tatham R.L., 2010. Multivariate data

analysis: a global perspective. 6th edition. Upper Saddle River: Pearson Education Inc.

Hall, E.T., 1981. Beyond Culture. New York: Anchor Books.

Harris, L.C. and Goode, M.M.H., 2004. The Four Levels of Loyalty and the Pivotal Role of Trust:

A Study of Online Service Dynamics. Journal of Retailing [online], 80 (2), 139-158.

Harrison, R.A., Holt, D. and Elton, P.J., 2002. Do postage-stamps increase response rates to

postal surveys? A randomized controlled trial. International Journal of Epidemiology [online],

31 (4), 872-874.

199

Hart, C.W. and Johnson, M.D., 1999. Growing the Trust Relationship. Marketing Management

[online], 8 (1), 8-19.

Hennig-Thurau, T., Gwinner, K.P. and Gremler, D.D., 2002. Understanding Relationship

Marketing Outcomes: An Integration of Relational Benefits and Relationship Quality. Journal of

Service Research [online], 4 (3), 230-247.

Hofstede, G., 1980. Culture’s Consequences: International Differences in Work-Related Values.

Newbury Park: SAGE Publications, Inc.

Hofstede, G., 1991. Culture’s Consequences. International Differences in Work-Related Values.

Newbury Park: SAGE Publications, Inc.

Hofstede, G. and Bond, M.H., 1988. The Confucius Connection: From Cultural Roots to

Economic Growth. Organizational Dynamics [online], 16 (4), 4-21.

Hofstede, G., 2002. The Pitfalls of Cross-National Survey Research: A Reply to the Article by

Spector et al. on the Psychometric Properties of the Hofstede Values Survey Module 1994.

Applied Psychology: An International Review [online], 51 (1), 170-174.

Hofstede, G.H., Hofstede, G.J. and Minkov, M., 2010. Culture and Organizations: Software of

the Mind: International Cooperation and its Importance for Survival. London: McGraw-Hill.

Hofstede Insights, 2018. Country Comparison. [online]. Available from: https://www.hofstede-

insights.com/country-comparison/sri-lanka,the-uk/ [Accessed 10 January 2018].

Hoft, N., 1996. Developing a Cultural Model. In: Del Galdo, E.M. and Nielsen, J., eds.

International User Interfaces. New York: John Wiley and Sons, Inc., 41-73.

House, R.J., Hanges, P.J., Ruiz-Quintanilla, S.A., Dorfman, P.W., Javidan, M., Dickson, M. and

Gupta, V., 1999. Cultural Influences on Leadership and Organizations: Project GLOBE. In:

Mobley, W.H., ed. Advances in Global Leadership, Volume 1. US: Elsevier Science/JAI Press,

171-233.

House, R.J., Javidan, M. and Dorfman, P.W., 2004. Culture, Leadership, and Organizations: The

GLOBE Study of 62 Societies. Thousand Oaks: Sage Publications, Inc.

Houston, M.J. and Jefferson, R.W., 1975. The Negative Effects of Personalization on Response

Patterns in Mail Surveys. Journal of Marketing Research [online], 12 (1), 114-117.

200

Howe, K.R., 1988. Against the Quantitative-Qualitative Incompatibility Thesis or Dogmas Die

Hard. Educational Researcher [online], 17 (8), 10-16.

Howell, R.D. and Rogers, J.D., 1981. Research into Shopping Mall Choice Behavior. Advances in


Hu, F.K. and Weber, K., 2014. The Influence of Culture on the Perceived Attractiveness of Hotel

Loyalty Programs: Chinese versus Japanese Customers. Journal of China Tourism Research

[online], 10 (2), 186-205.

Huang, M.H., 2015. The influence of relationship marketing investments on customer gratitude

in retailing. Journal of Business Research [online], 68 (6), 1318-1323.

Huppertz, J.W., Arenson, S.J. and Evans, R.H., 1978. An Application of Equity Theory to Buyer-

Seller Exchange Situations. Journal of Marketing [online], 15 (2), 250-60.

IGD Retail Analysis, 2013. 2013 Christmas Strategies: What You Should Know [online].

Watforfd: The Institute of Grocery Distribution and IGD Services. Available from:

https://retailanalysis-igd-com.libezproxy.bournemouth.ac.uk/markets/presentation-

viewer?t=christmas-trading-2013-uk-festive-trading-outlook&i=1174 [Accessed 13 January

2018].

IGD Retail Analysis, 2015. Christmas 2015: Festive Trading Outlook [online]. Watforfd: The

Institute of Grocery Distribution and IGD Services. Available from: https://retailanalysis-igd-

com.libezproxy.bournemouth.ac.uk/markets/presentation-viewer?t=christmas-2015-uk-

festive-trading-outlook&i=1795 [Accessed 13 January 2018].

IGD Retail Analysis, 2017a. UK Country Presentation [online]. Watforfd: The Institute of

Grocery Distribution and IGD Services. Available from:

https://retailanalysis.igd.com/markets/presentation-viewer?t=uk-country-

presentation&i=7112 [Accessed 11 January 2018].

IGD Retail Analysis, 2017b. UK Channel Opportunities: 2017-2022 [online]. Watforfd: The

Institute of Grocery Distribution and IGD Services. Available from:

https://retailanalysis.igd.com/markets/presentation-viewer?t=uk-channel-opportunities-2017-

2022&i=6910 [Accessed 11 January 2018].

IGD Retail Analysis, 2017c. Tesco Enters Pakistan with Private Label Deal [online]. Watforfd:

The Institute of Grocery Distribution and IGD Services. Available from:

201

http://retailanalysis.igd.com.libezproxy.bournemouth.ac.uk/Hub.aspx?id=23&tid=3&nid=1629

5 [Accessed 14 January 2018].

IGD Shopper Vista, 2016. Understanding Shoppers at Christmas [online]. Watforfd: The

Institute of Grocery Distribution and IGD Services. Available from: https://shoppervista-igd-

com.libezproxy.bournemouth.ac.uk/trends/presentation-viewer/t/understanding-shoppers-at-

christmas/i/667 [Accessed 13 January 2018].

IGD Shopper Vista, 2017a. Retail and Shopper Trends: Q3 2017 [online]. Watforfd: The Institute

of Grocery Distribution and IGD Services. Available from: https://shoppervista-igd-

com.libezproxy.bournemouth.ac.uk/trends/presentation-viewer/t/retail-and-shopper-trends-

q3-2017/i/7234 [Accessed 11 January 2018].

IGD Shopper Vista, 2017b. The role of brands in supermarkets [online]. Watforfd: The Institute

of Grocery Distribution and IGD Services. Available from: https://shoppervista-igd-

com.libezproxy.bournemouth.ac.uk/trends/presentation-viewer?t=the-role-of-brands-in-

supermarkets&i=7123 [Accessed 11 January 2018].

Imran, M., Ghani, U. and Rehman, K.U., 2013. Consumer Perception of Store Image and Store

Loyalty. Journal of Managerial Sciences [online], 7, 75-85.

Institute of Race Relations, 2018. Ethnicity and Religion Statistics [online]. London: Institute of

Race Relations. Available from: http://www.irr.org.uk/research/statistics/ethnicity-and-

religion/ [Accessed 12 January 2018].

International Business Times, 2015. UK shoppers prefer Aldi and Lidl for Christmas groceries

[online]. London: IBTimes Co., Ltd. Available from: http://www.ibtimes.co.uk/uk-shoppers-

prefer-aldi-lidl-christmas-groceries-1531407 [Accessed 13 January 2018].

Jacoby, J., 1971. Brand Loyalty: A Conceptual Definition. Proceedings of the Annual Convention

of the American Psychological Association [online], 6 (Part 2), 655-656.

Jacoby, J. and Matell, M.S., 1971. Three-Point Likert Scales Are Good Enough. Journal of

Marketing Research [online], 8 (4), 495-500.

Jacoby, J. and Kyner, D.B., 1973. Brand Loyalty Vs. Repeat Purchasing Behavior. Journal of

Marketing Research (JMR) [online], 10, 1-9.

Jamieson, S., 2004. Likert scales: how to (ab)use them. Medical Education [online], 38 (12),

1217-1218.

202

Jang, D. and Mattila, A.S., 2005. An examination of restaurant loyalty programs: what kinds of

rewards do customers prefer? International Journal of Contemporary Hospitality Management

[online], 17 (5), 402-408.

Javidan, M., House, R.J., Dorfman, P.W., Hanges, P.J. and Luque, M.S.D., 2006. Conceptualizing

and Measuring Cultures and their Consequences: A Comparative Review of GLOBE’s and

Hofstede’s Approaches. Journal of International Business Studies [online], 37, 897-914.

Jensen, J.M., 2011. Consumer Loyalty on the Grocery Product Market: An Empirical Application

of Dick and Basu's Framework. Journal of Consumer Marketing [online], 28 (5), 333-343.

Johnson, B. and Turner, L.A., 2003. Data Collection Strategies in Mixed Methods Research. In:

Tashakkori, A. and Teddlie, C., eds. SAGE Handbook of Mixed Methods in Social and Behavioral

Research. Thousand Oaks: SAGE Publications, Inc., 297-319.

Johnson, R.B. and Onwuegbuzie, A.J., 2004. Mixed Methods Research: A Research Paradigm

Whose Time Has Come. Educational Researcher [online], 33 (7), 14-26.

Johnson, R.B., Onwuegbuzie, A.J. and Turner, L.A., 2007. Toward a Definition of Mixed

Methods Research. Journal of Mixed Methods Research [online], 1 (2), 112-133.

Kantar Worldpanel, 2018. Grocery Market Share (12 weeks ending) [online]. London: Kantar

UK. Available from: https://www.kantarworldpanel.com/en/grocery-market-share/great-

britain/snapshot/31.12.17/01.01.17 [Accessed 14 January 2018].

Kaul, S., 2007. Hedonism and Culture: Impact on Shopping Behaviour a Research Agenda.

Vikalpa: The Journal for Decision Makers [online], 32 (3), 81-89.

Keells Super, 2018. Keells Super [online]. Colombo: John Keells Holdings PLC. Available from:

https://www.keellssuper.com/contact [Accessed 11 January 2018].

Keh, H.T. and Lee, Y.H., 2006. Do reward programs build loyalty for services? The moderating

effect of satisfaction on type and timing of rewards. Journal of Retailing [online], 82 (2), 127-

136.

Key Note, 2015. Supermarket Own Labels: Food & Drink [online]. London: Key Note Ltd.

Available from: https://www.keynote.co.uk/market-report/retail/supermarket-own-labels-

food-drink [Accessed 11 January 2018].

203

Khare, A., 2012. Influence of Culture on Indian Consumers’ Preference to Shop at Small Retail

Stores. Journal of Global Marketing [online], 25 (2), 100-111.

Khare, A., Pandey, S.K. and Bhardwaj, P., 2014. Impact of Culture, Cosmopolitanism, and Price

on Local Store Loyalty: An Empirical Study from India. Journal of International Consumer


King, M.F. and Bruner, G.C., 2000. Social Desirability Bias: A Neglected Aspect of Validity

Testing. Psychology & Marketing [online], 17 (2), 79-103.

Kirkman, B.L., Lowe, K.B. and Gibson, C.B., 2006. A Quarter Century of Culture’s Consequences:

A Review of Empirical Research Incorporating Hofstede’s Cultural Values Framework. Journal

of International Business Studies [online], 37 (3), 285-320.

Kitapci, O., Dortyol, I.T., Yaman, Z. and Gulmez, M., 2013. The Paths from Service Quality

Dimensions to Customer Loyalty. An Application on Supermarket Customers. Management

Research Review [online], 36(3), 239-255.

Kitzinger, J., 1995. Introducing Focus Groups. British Medical Journal [online], 311 (7000), 299-

302.

Klausch, T., Schouten, B. and Hox, J.J., 2017. Evaluating Bias of Sequential Mixed-mode Designs

Against Benchmark Surveys. Sociological Methods & Research [online], 46 (3), 456-489.

Kluckhohn, F. and Strodtbeck, F., 1961. Variations in Value Orientations. Evanston, IL: Row,

Peterson.

Knox, S.D., 1998. Loyalty-Based Segmentation and the Customer Development Process.

European Management Journal [online], 16 (6), 729-737.

Knox, S.D. and Denison, T.J., 2000. Store Loyalty: Its Impact on Retail Revenue. An Empirical

Study of Purchasing Behaviour in the UK. Journal of Retailing and Consumer Services [online],

7, 33-45.

Koschate-Fischer, N., Cramer, J. and Hoyer, W.D., 2014. Moderating Effects of the Relationship

Between Private Label Share and Store Loyalty. Journal of Marketing [online], 78 (2), 69-82.

Kroeber, A.L. and Parsons, T., 1958. The Concepts of Culture and of Social System. American

Sociological Review [online], 23 (5), 582-583.

Krueger, R.A., 1998. Moderating Focus Groups. London: SAGE Publications, Inc.

204

Krueger, R.A. and Casey, M.A., 2009. Focus Groups: A Practical Guide for Applied Research. 4th

edition. London: SAGE Publications, Inc.

Krueger, R.A., and Casey, M.A., 2015. Focus Groups: A Practical Guide for Applied Research. 5th

edition. California: SAGE Publications, Inc.

Kueh, K. and Voon, B.H., 2007. Culture and service quality expectations. Evidence from

Generation Y consumers in Malaysia. Managing Service Quality [online], 17 (6), 656-680.

Kumar, V. and Shah, D., 2004. Building and Sustaining Profitable Customer Loyalty for the 21st

Century. Journal of Retailing [online], 80 (4), 317-330.

Laaksonen, M., 1993. Retail Patronage Dynamics: Learning about Daily Shopping Behavior in

Contexts of Changing Retail Structures. Journal of Business Research [online], 28 (1/2), 3-174.

Lacey, R., Suh, J. and Morgan, R.M., 2007. Differential effects of preferential treatment levels

on relational outcomes. Journal of Service Research [online], 9 (3), 241-256.

Lam, D.C.S., 2007. Cultural Influence on Proneness to Brand Loyalty. Journal of International

Consumer Marketing [online], 19 (3), 7-21.

Laroche, M., Ueltschy, L.C., Abe, S., Cleveland, M. and Yannopoulos, P.P., 2004. Service Quality

Perceptions and Customer Satisfaction: Evaluating the Role of Culture. Journal of International


Laugfs Holdings, 2018. Locations [online]. Colombo: Laugfs Holdings Ltd. Available from:

https://laugfsholdings.com/sunup/location.htm [Accessed 11 January 2018].

Lavelle, K., Todd, C. and Campbell, M., 2008. Do postage stamps versus pre-paid envelopes

increase responses to patient mail surveys? A randomised controlled trial. BMC Health Services

Research [online], 8 (113), 1-5.

Lawal, M., 2009. Reconciling Methodological Approaches of Survey and Focus Group. Nurse

Researcher [online], 17 (1), 54-61.

Leenheer, J., van Heerde, H.J., Bijmolt, T.H.A. and Smidts, A., 2007. Do Loyalty Programs Really

Enhance Behavioral Loyalty? An Empirical Analysis Accounting for Self-Selecting Members.

International Journal of Research in Marketing [online], 24, 31-47.

205

Leppäniemi, M., Karjaluoto, H. and Saarijärvi, H., 2017. Customer perceived value, satisfaction,

and loyalty: the role of willingness to share information. The International Review of Retail,

Distribution and Consumer Research [online], 27 (2), 164-188.

Lessig, V.P., 1973. Consumer Store Images and Store Loyalties. Journal of Marketing [online],

37 (4), 72-74.

Leung, S., 2011. A Comparison of Psychometric Properties and Normality in 4-, 5-, 6-, and 11-

Point Likert Scales. Journal of Social Service Research [online], 37 (4), 412-421.

Levy, R.M., Shapiro, M., Halpern, S.D. and Ming, M.E., 2012. Effect of Personalization and

Candy Incentive on Response Rates for a Mailed Survey of Dermatologists. Journal of

Investigative Dermatology [online], 132 (3), 724-726.

Lewis M., 2004. The influence of loyalty programs and short-term promotions on customer

retention. Journal of Marketing Research [online], 41 (3), 281-292.

Likert, R., 1932. A technique for the measurement of measurements. Archives of Psychology

[online], 22 (140), 55.

Lindquist, J.D., 1974. Meaning of Image: Survey of Empirical and Hypothetical Evidence.

Journal of Retailing [online], 50 (4), 29-38.

Lindridge, A., Vijaygopal, R. and Dibb, S., 2014. The Manifestation of Culture in Product

Purchase: A Cross-Cultural Comparison. Journal of Marketing Analytics [online], 2 (4), 250-263.

Little, R.J.A., 1988. A Test of Missing Completely at Random for Multivariate Data with Missing

Values. Journal of the American Statistical Association [online], 83 (404), 1198-1202.

Liu, Y., 2009. The Long-Term Impact of Loyalty Programs on Consumer Purchase Behavior and

Loyalty. Journal of Marketing [online], 71 (4), 19-35.

LMD, 2016. RETAIL IS KING [online]. Colombo: Media Services (Private) Limited. Available from:

https://lmd.lk/retail-is-king/ [Accessed 11 January 2018].

LMD, 2017. Spotlight: Sri Lankan Modern Grocery Retail [online]. Colombo: Media Services

(Private) Limited. Available from: https://lmd.lk/wp-content/uploads/2017/11/Spotlight-Sri-

Lankas-modern-grocery-retail-Nov-2017.pdf [Accessed 15 January 2018].

LMRB, 2016. LMRB Consumer Pulse. Colombo: Lanka Market Research Bureau.

206

Luceri, B. and Latusi, S., 2012. The Importance of Consumer Characteristics and Market

Structure Variables in Driving Multiple Store Patronage. Journal of Retailing and Consumer

Services [online], 19, 519-525.

Lugtig, P., Lensvelt-Mulders, G.J.L.M., Frerichs, R. and Greven, A., 2011. Estimating

nonresponse bias and mode effects in a mixed-mode survey. International Journal of Market


Luna, D. and Gupta, S.F., 2001. An Integrative Framework for Cross-Cultural Consumer

Behavior. International Marketing Review [online], 18, 45-69.

Lund, E. and Gram, I.T., 1988. Response rate according to title and length of questionnaire.

Scandinavian Journal of Social Medicine [online], 26 (2), 154-160.

MacKenzie, S.B. and Podsakoff, P.M., 2012. Common method bias in marketing: Causes,

mechanisms, and procedural remedies. Journal of Retailing [online], 88 (4), 542-555.

Magnusson, P., Wilson, R.T., Zdravkovic, S., Zhou, J.X. and Westjohn, S.A., 2008. Breaking

through the cultural clutter. A comparative assessment of multiple cultural and institutional

frameworks. International Marketing Review [online], 25 (2), 183-201.

Malhotra, N.K. and Birks, D.F., 2005. Marketing Research: An Applied Approach. 2nd European

edition. Harlow: Financial Times Prentice Hall.

Malhotra, N.K. and Dash, S., 2011. Marketing Research: An Applied Approach. Harlow: Pearson

Education Limited.

Mango, 2018. Mango [online]. Colombo: Mango. Available from:

https://shop.mango.com/lk/women [Accessed 11 January 2018].

Masters, C., Carlson, D.S. and Pfadt, E., 2006. Winning it through Research: An Innovative

Approach to a Basic Understanding of Research Methodology. Journal of Emergency Nursing

[online], 32 (5), 382-384.

Martineau, P., 1958. The Personality of the Retail Store. Harvard Business Review [online], 36,

47-55.

Martinelli, E. and Balboni, B., 2012. Retail Service Quality as a Key Activator of Grocery Store

Loyalty. The Service Industries Journal [online], 32 (14), 2233-2247.

207

Maruyama, M. and Wu, L., 2014. Multiple store patronage: The effects of store characteristics.

Journal of Retailing and Consumer Services [online], 21 (4), 601-609.

Mattila, A.S., 2000. The Impact of Culture and Gender on Customer Evaluations of Service

Encounters. Journal of Hospitality & Tourism Research [online], 24 (2), 263-273.

Mattison Thompson, F. and Chmura, T., 2015. Loyalty Programs in Emerging and Developed

Markets: The Impact of Cultural Values on Loyalty Program Choice. Journal of Interational


Mauri, C., 2003. Card Loyalty. A New Emerging Issue in Grocery Retailing. Journal of Retailing &

Consumer Services [online], 10, 13-25.

Mazaheri, E., Richard, M.O., Laroche, M. and Ueltschy, L.C., 2014. The Influence of Culture,

Emotions, Intangibility, and Atmospheric Cues on ONline Behavior. Journal of Business


Meade, A.W. and Bauer, D.J., 2007. Power and Precision in Confirmatory Factor Analytic Tests

of Measurement Invariance. Structural Equation Modeling [online], 14 (4), 611-635.

Meyer-Waarden, L. 2008. The influence of loyalty programme membership on customer

purchase behaviour. European Journal of Marketing [online], 42 (1/2), 87-114.

Meyer-Waarden, L. and Benavent, C., 2009. Grocery Retail Loyalty Program Effects: Self-

Selection or Purchase Behavior Change? Journal of the Academy of Marketing Science [online],

37 (3), 345-358.

Meyer-Waarden, L. 2015. Effects of loyalty program rewards on store loyalty. Journal of


Mimouni-Chaabane, A. and Volle, P., 2010. Perceived benefits of loyalty programs: Scale

development and implications for relational strategies. Journal of Business Research [online],

63 (1), 32-37.

Minnema, A., Bijmolt, T.H.A. and Non, M.C., 2017. The impact of instant reward programs and

bonus premiums on consumer purchase behavior. International Journal of Research in


Mintel, 2016a. Supermarkets - UK - November 2016 [online]. London: Mintel Group Ltd.

Available from: http://academic.mintel.com/display/804109/ [Accessed 11 January 2018].

208

Mintel, 2016b. Supermarkets - UK - November 2016 [online]. London: Mintel Group Ltd.


Mintel, 2016c. Supermarkets - UK - November 2016 [online]. London: Mintel Group Ltd.


Mintel, 2017a. Online Grocery Retailing - UK - March 2017 [online]. London: Mintel Group Ltd.


Mintel, 2017b. Supermarkets - UK - November 2017 [online]. London: Mintel Group Ltd.


Mintel, 2017c. Specialist Food & Drink Retailers - UK - March 2017 [online]. London: Mintel

Group Ltd. Available from: http://academic.mintel.com/display/825449/ [Accessed 13 January

2018].

Molina, A., Martin, V.J., Santos J. and Aranda E., 2009. Consumer Service and Loyalty in Spanish

Grocery Store Retailing: An Empirical Study. International Journal of Consumer Studies [online],

33 (4), 477-485.

Moorman, C., Deshpande, R. and Zaltman, G., 1993. Relationships Between Providers and

Users of Market Research: The Role of Personal Trust. Report- Marketing Science Institute

Cambridge Massachusetts [online], 111, All.

Morgan, D.L. and Spanish, M.T., 1984. Focus Groups: A New Tool for Qualitative Research.

Qualitative Sociology [online], 7 (3), 253-271.

Morgan, D.L., 1997. Focus Groups as Qualitative Research. 2nd edition. London: SAGE

Publications, Inc.

Morgan, D.L., 1998. The Focus Group Guidebook. 1st edition. London: SAGE Publications, Inc.

Morgan, D.L. and Scannell, A.U., 1998. Planning Focus Groups. London: SAGE Publications, Inc.

Morgan, R.M. and Hunt, S.D., 1994. The Commitment-Trust Theory of Relationship Marketing.

Journal of Marketing [online], 58 (3), 20-38.

Nakash, R.A., Hutton, J.L., Jorstad-Stein, E.C., Gates, S. and Lamb, S.E., 2006. Maximising

response to postal questionnaires - A systematic review of randomised trials in health

research. BMC Medical Research Methodology [online], 6 (5), 1-9.

209

Nakata, C. and Sivakumar, K., 2001. Instituting the Marketing Concept in a Multinational

Setting: The Role of National Culture. Journal of the Academy of Marketing Science [online], 29

(3), 255-275.

Nardon, L. and Steers, R.M., 2006. Navigating the Culture Theory Jungle: Divergence and

Convergence in Models of National Culture. Vlerick Leuven Gent Working Paper Series [online],

38, 1-43.

Ndubisi, N.O., Malhotra, N.K., Ulas, D. and Ndubisi, G.C., 2012. Examining Uncertainty

Avoidance, Relationship Quality, and Customer Loyalty in Two Cultures. Journal of

International Consumer Marketing [online], 24 (5), 320-337.

Nesset, E., Nervik, B. and Helgesen, O., 2011. Satisfaction and Image as Mediators of Store

Loyalty Drivers in Grocery Retailing. The International Review of Retail, Distribution and


Ngobo, P.V., 2017. The trajectory of customer loyalty: an empirical test of Dick and Basus’s

loyalty framework. Journal of the Academy of Marketing Science [online], 45 (2), 229-250.

Nguyen, B. and Mutum, D.S., 2012. A Review of Customer Relationship Management:

Successes, Advances, Pitfalls and Futures. Business Process Management Journal [online], 18

(3), 400-419.

Nies, S and Natter, M., 2012. Does Private Label Quality Influence Consumers’ Decision on

Where to Shop? Psychology & Marketing [online], 29 (4), 279-292.

Noordhoff, C., Pauwels, P. and Odekerken-Schroder, G., 2004. The Effect of Customer Card

Programs. A Comparative Study in Singapore and The Netherlands. International Journal of

Service Industry Management [online], 15 (4), 351-364.

Norberg, P.A., Horne, D.R. and Horne, D.A., 2007. The Privacy Paradox: Personal Information

Disclosure Intentions versus Behaviors. The Journal of Consumer Affairs [online], 41 (1), 100-

126.

Nunnally, J., 1978. Psychometric Theory. McGraw-Hill: New York.

O'Brien, K., 1993. Using Focus Groups to Develop Health Surveys: An Example from Research

on Social Relationships and AIDS-Preventive Behavior. Health Education & Behavior [online],

20 (3), 361-372.

210

Odekerken-Schroder, G., De Wulf, K., Kasper, H., Kleijnen, M., Hoekstra, J. and Commandeur

H., 2001. The Impact of Quality on Store Loyalty: A Contingency Approach. Total Quality

Management [online], 12 (3), 307-322.

Office for National Statistics, 2017a. Retail sales pounds data [online]. London: Office for

National Statistics. Available from:

https://www.ons.gov.uk/businessindustryandtrade/retailindustry/datasets/poundsdatatotalre

tailsales [Accessed 11 January 2018].

Office for National Statistics, 2017b. Population Estimates for UK, England and Wales, Scotland

and Northern Ireland [online]. London: Office for National Statistics. Available from:

https://www.ons.gov.uk/businessindustryandtrade/retailindustry/datasets/poundsdatatotalre

tailsales [Accessed 12 January 2018].

Office for National Statistics, 2017c. Migration Statistics Quarterly Report: November 2017

[online]. London: Office for National Statistics. Available from:

https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/internatio

nalmigration/bulletins/migrationstatisticsquarterlyreport/november2017 [Accessed 14

January 2018].

Office for National Statistics, 2018. Household disposable income and inequality [online].

London: Office for National Statistics. Available from:

https://www.ons.gov.uk/peoplepopulationandcommunity/personalandhouseholdfinances/inc

omeandwealth/datasets/householddisposableincomeandinequality [Accessed 12 January

2018].

Oliver, R.L., 1997. Satisfaction: A Behavioural Perspective on the Consumer. New York:

Irwin/McGraw-Hill.

Oliver, R.L., 1999. Whence Consumer Loyalty? Journal of Marketing [online], 63 (4), 33-44.

Omar, O. and Sawmong, S., 2007. Customer Satisfaction and Loyalty to British Supermarkets.

Journal of Food Products Marketing, 13 (2), 19-32.

Omar, N.A., Wel, C.A.C., Musa, R. and Nazri, M.A., 2010. Program Benefits, Satisfaction and

Loyalty in Retail Loyalty Program: Exploring the Roles of Program Trust and Program

Commitment. The IUP Journal of Marketing Management [online], 9 (4), 6-28.

211

Omar, N.A., Aziz, N.A. and Nazri, M.A., 2011. Understanding the Relationships of Program

Satisfaction, Program Loyalty and Store Loyalty among Cardholders of Loyalty Programs. Asian

Academy of Management Journal [online], 16 (1), 21-41.

Omar, N.A., Wel, C.A.C., Aziz, N.A. and Alam, S.S., 2013. Investigating the structural

relationship between loyalty programme service quality, satisfaction and loyalty for retail

loyalty programmes: evidence from Malaysia. Measuring Business Excellence [online], 17 (1),

33-50.

Oppenheim, A.N., 1992. Questionnaire Design, Interviewing and Attitude Measurement. New

edition. London: Continuum.

Osland, J.S. and Bird, A., 2000. Beyond Sophisticated Stereotyping: Cultural Sensemaking in

Context. The Academy of Management Executive [online], 14, 65-79.

Ou, W.M., Shih, C.M. and Chen, C.Y., 2014. Antecedents and consequences of relationship

commitment: an empirical study in Taiwan. International Journal of Commerce and

Management [online], 24 (3), 228-242.

Oxford English Dictionary, 2018. Culture [online]. Oxford: Oxford University Press. Available

from: http://www.oed.com/view/Entry/45746?rskey=8E4ep4&result=1&isAdvanced=false#eid


Pallant, J., 2010. SPSS survival manual: a step by step guide to data analysis using SPSS. 4th

edition. Maidenhead: McGraw Hill.

Pallant, J., 2016. SPSS survival manual: a step by step guide to data analysis using SPSS. 6th

edition. Maidenhead: McGraw Hill.

Palmer, A. J., 1996. Integrating Brand Development and Relationship Marketing. Journal of


Palmer, A., McMohan-Beattie, U. and Beggs, R., 2000. Influences on loyalty programme

effectiveness: a conceptual framework and case study investigation. Journal of Strategic

Marketing [online], 8, 47-66.

Palumbo, F. and Herbig, P., 2000. The Multicultural Context of Brand Loyalty. European Journal

of Innovation Management [online], 3 (3), 116-125.

212

Pandey, S., Khare, A. and Bhardwaj, P., 2015. Antecedents to Local Store Loyalty: Influence of

Culture, Cosmopolitanism and Price. International Journal of Retail & Distribution

Management [online], 43, 5-25.

Parasuraman, A., Zeithaml, V.A. and Berry, L.L., 1985. A Conceptual Model of Service Quality

and Implications for Future Research. Journal of Marketing [online], 49, 41-50.

Park, S.B., Chung, N. and Woo, S.C., 2013. Do Reward Programs Build Loyalty To Restaurants?

The Moderating Effect Of Long-Term Orientation on the Timing and Types of Rewards.

Managing Service Quality: An International Journal [online], 23 (3), 225-244.

Patton, M.Q., 2002. Qualitative Research and Evaluation Methods. 3rd edition. London: SAGE

Publications, Inc.

Pearse, N., 2011. Deciding on the Scale Granularity of Response Categories of Likert type

Scales: The Case of a 21-Point Scale. The Electronic Journal of Business Research Methods

[online], 9 (2), 159-171.

Pepe, M.S., Abratt, R. and Dion, P., 2011. The impact of private label brands on customer

loyalty and product category profitability. Journal of Product & Brand Management [online], 20

(1), 27-36.

Peppers, D., Rogers, M., Reichheld, F., Reinartz, W. and Kumar, V., 2002. The Mismanagement

of Customer Loyalty. Harvard Business Review [online], 80 (11), 125-129.

Perneger, T.V., Courvoisier, D.S., Hudelson, P.M. and Gayet-Ageron, A., 2015. Sample size for

pre-tests of questionnaires. Quality of Life Research [online], 24 (1), 147-151.

Pez, V., Butori, R. and Mimouni-Chaabane, A., 2017. The dark side of the pressure exerted

by loyalty programs on consumers: Practical and ethical issues. Recherche et Applications en


Podsakoff, P.M. and Organ, D.W., 1986. Self-Reports in Organizational Research: Problems and

Prospects. Journal of Management [online], 12 (4), 531-544.

Podsakoff, P.M., MacKenzie, S.B., Lee J. and Podsakoff, N.P., 2003. Common method biases in

behavioral research: A critical review of the literature and recommended remedies. Journal of

Applied Psychology [online], 88 (5), 879-903.

213

Poole, R., Clarke, G.P. and Clarke, D.B., 2006. Competition and saturation in West European

grocery retailing. Environment and Planning [online], 38 (11), 2129-2156.

Pyke, S.J., 2017. A systems theory approach to the well-being effects of tourism in the United

Kingdom [online]. Thesis (PhD). Bournemouth University.

Ramaseshan, B. and Ouschan, R., 2017. Investigating status demotion in hierarchical loyalty

programs. Journal of Services Marketing [online], 31 (6), 650-661.

Ranaweera, C. and Neely, A., 2003. Some moderating effects on the service quality-customer

retention link. International Journal of Operations & Production Management [online], 23 (2),

230-248.

Reardon, T., Timmer, C.P. and Minten, B., 2012. Supermarket revolution in Asia and emerging

development strategies to include small farmers. Proceedings of the National Academy of

Sciences of the United States of America [online], 109 (31), 12332-12337.

Reichheld, F.R. and Sasser Jr, W.E., 1990. Zero defections: Quality Comes to Services. Harvard

Business Review [online], 68 (5), 105-111.

Reichheld, F.R., 1993. Loyalty-Based Management. Harvard Business Review [online], 71 (2),

64-73.

Reichheld, F.R., 1994. Loyalty and the Renaissance of Marketing. Marketing Management

[online], 2 (4), 10-21.

Reichheld, F.R., Markey Jr., R.G. and Hopton, C., 2000. The Loyalty Effect- The Relationship

between Loyalty and Profits. European Business Journal [online], 12 (3), 134-139.

Retrowow, 2015. Green Shield Stamps [online]. Unknown Location. Available from:

http://www.retrowow.co.uk/retro_britain/green_shield_stamps/Green_Shield_Stamps.html


Richardson, H.A., Simmering, M.J. and Sturman M.C., 2009. A Tale of Three Perspectives:

Examining Post Hoc Statistical Techniques for Detection and Correction of Common Method

Variance. Organizational Research Methods [online], 12 (4), 762-800.

Rowley, J., 2005. The Four Cs of Customer Loyalty. Marketing Intelligence & Planning [online],

23 (6), 574-581.

214

Royal Mail, 2018. Postcode Finder – Find an Address [online]. London: Royal Mail Group Ltd.

Available from: https://www.royalmail.com/find-a-postcode [Accessed 12 January 2018].

Saegert, J., Hoover, R.J. and Hilger, M.T., 1985. Characteristics of Mexican American

Consumers. Journal of Consumer Research [online], 12, 104-109.

Sahlqvist, S., Song, Y., Bull, F., Adams, E., Preston, J., Ogilvie, D. and for the iConnect

consortium., 2011. Effect of questionnaire length, personalisation and reminder type on

response rate to a complex postal survey: randomised controlled trial. BMC Medical Research

Methodology [online], 11 (62), 1-8.

Sala, E. and Lynn, P., 2009. The potential of a multi-mode data collection design to reduce non

response bias. The case of a survey of employers. Quality & Quantity: International Journal of

Methodology [online], 43 (1), 123-136.

Saunders, M., Lewis, P., and Thornhill, A., 2012. Research Methods for Business Students. 6th

edition. Harlow: Pearson Education Limited.

Sawmong, S. and Omar, O., 2004. The Store Loyalty of the UK's Retail Consumers. Journal of

American Academy of Business, Cambridge [online], 5 (1/2), 503-509.

Sawmong, S., 2006. A Comparitive Study of Grocery Store Loyalty in the United Kingdom and

Thailand [online]. Thesis (PhD). London South Bank University.

Schirmer, J., 2009. Ethical Issues in the Use of Multiple Survey Reminders. Journal of Academic

Ethics [online], 7 (1/2), 125-139.

Schwartz, S.H., 1999. A Theory of Cultural Values and Some Implications for Work. Applied

Psychology: An International Review [online], 48, 23-47.

Schmitt, B.H. and Pan, Y., 1994. Managing corporate and brand identities in the Asia-Pacific

region. California: University of California Press.

Seiders, K. and Tigert, D.J., 1997. Impact of Market Entry and Competitive Structure on Store

Switching/Store Loyalty. The International Review of Retail, Distribution and Consumer


Seock, Y.K. and Lin, C., 2011. Cultural Influence on Loyalty Tendency and Evaluation of Retail

Store Attributes: An Analysis of Taiwanese and American Consumers. International Journal of

Retail and Distribution Management [online], 39 (2), 94-113.

215

Shannon, R. and Mandhachitara, R., 2005. Private-label grocery shopping attitudes and

behaviour: A cross-cultural study. Brand Management [online], 12 (6), 461-474.

Sharp, B. and Sharp, A., 1997. Loyalty Programs and their Impact on Repeat-Purchase Loyalty

Patterns. International Journal of Research in Marketing [online], 14 (5), 473-486.

Sheth, J. N. and Parvatiyar, A., 1995. Paper: The Evolution of Relationship Marketing.

International Business Review [online], 4, 397-418.

Shi, X. and Wang, J., 2011. Interpreting Hofstede Model and GLOBE Model: Which Way to go

for Cross-Cultural Research? International Journal of Business and Management [online], 6 (5),

93-99.

Shoemaker, S. and Lewis, R.C., 1999. Customer Loyalty: The Future of Hospitality Marketing.

International Journal of Hospitality Management [online], 18 (4), 345-370.

Shpetim, C., 2012. Exploring the Relationships among Service Quality, Satisfaction, Trust and

Store Loyalty among Retail Customers. Journal of Competitiveness [online], 4 (4), 16-35.

Siegfried, F., 2014. A Cross-Cultural Study of Consumer Switching in the Retail Banking Service

Sector [online], Thesis (PhD). University of Manchester.

Simmons, R., 2008. Questionnaires. In: Gilbert, N., ed. Researching Social Life. 3rd edition.

London: SAGE Publications Inc., 85-104.

Sirohi, N., McLaughlin, E.W. and Wittink, D.R., 1998. A Model of Consumer Perceptions and

Store Loyalty Intentions for a Supermarket Retailer. Journal of Retailing [online], 74 (2), 223-

245.

Sivadas, E. and Baker-Prewitt, J.L., 2000. An Examination of the Relationship between Service

Quality, Customer Satisfaction, and Store Loyalty. International Journal of Retail and

Distribution Management [online], 28 (2), 73-82.

Smith, J.B. and Barclay, D.W., 1997. The Effects of Organizational Differences and Trust on the

Effectiveness of Selling Partner Relationships. Journal of Marketing [online], 61 (1), 3-21.

Smith, J.B., 1998. Buyer-Seller Relationships: Bonds, Relationship Management, and Sex-Type.

Canadian Journal of Administrative Sciences [online], 15 (1), 76-92.

Smith, J.B., 1998. Buyer-Seller Relationships: Similarity, Relationship Management and Quality.

Psychology and Marketing [online], 15 (1), 3-21.

216

Smith, P.B., 2006. When Elephants Fight, The Grass Gets Trampled: The Globe and Hofstede

Projects. Journal of International Business Studies [online], 37 (6), 915-921.

Sojka, J.Z. and Tansuhaj, P.S., 1995. Cross-Cultural Consumer Research: A Twenty-Year Review.

Advances in Consumer Research [online], 22, 461-474.

Sondergaard, M., 1994. Research Note: Hofstede’s Consequences: A Study of Reviews,

Citations and Replications. Organization Studies [online], 15 (3), 447-456.

Spector, P.E., Cooper, C.L. and Sparks, K., 2001. An International Study of the Psychometric

Properties of the Hofstede Values Survey Module 1994: A Comparison of Individual and

Country/Province Level Results. Applied Psychology: An International Review [online], 50 (2),

269-281.

Spector, P.E. and Cooper, C.L., 2002. The Pitfalls of Poor Psychometric Properties: A Rejoinder

to Hofstede's Reply to Us. Applied Psychology: An International Review [online], 51, 174-178.

Sreedhara, R. and Babu, K.N., 2010. Factors Influencing Store Loyalty- A Conceptual Approach

on Supermarkets. Global Management Review [online], 4 (2), 37-42.

Steenkamp, J.E.M. and Baumgartner, H., 1998. Assessing Measurement Invariance in Cross-

National Consumer Research. Journal of Consumer Research [online], 25 (1), 78-90.

Steinhoff, L. and Palmatier, R.W., 2016. Understanding Loyalty Program Effectiveness:

Managing Target and Bystander Effects. Journal of the Academy of Marketing Science [online],

44 (1), 88-107.

Straughan, R.D. and Albers-Miller, N.D., 2001. An International Investigation of Cultural and

Demographic Effects on Domestic Retail Loyalty. International Marketing Review [online], 18

(5), 521-541.

Tanford, S., 2013. The impact of tier level on attitudinal and behavioral loyalty of hotel reward

program members. International Journal of Hospitality Management [online], 34 (2013), 285-

294.

Tashakkori, A. and Teddlie, C., 2003. SAGE Handbook of Mixed Methods in Social and

Behavioral Research. Thousand Oaks: SAGE Publications, Inc.

Tashakkori, A. and Creswell, J. W., 2007. The New Era of Mixed Methods. Journal of Mixed

Methods Research [online], 1, 3-7.

217

Tate, R.S., 1961. The Supermarket Battle for Store Loyalty. Journal of Marketing [online], 25

(6), 8-13.

Taylor, J.F., Ferguson, J. and Ellen, P.S., 2015. From trait to state: understanding privacy

concerns. Journal of Consumer Marketing [online], 32 (2), 99-112.

Teddlie, C. and Tashakkori, A., 2009. Foundations of Mixed Methods Research: Integrating

Quantitative and Qualitative Approaches in the Social and Behavioral Sciences. London: SAGE

Publications, Inc.

Teddlie, C. and Tashakkori, A., 2010. SAGE Handbook of Mixed Methods in Social and

Behavioral Research. 2nd edition. London: SAGE Publications, Inc.

Teller, C., Gittenberger, E. and Schnedlitz, P., 2013. Cognitive age and grocery-store patronage

by elderly shoppers. Journal of Marketing Management [online], 29 (3/4), 317-337.

Tesfom, G. and Birch, N.F., 2011. Do switching barriers in the retail banking industry influence

bank customers in different age groups differently? Journal of Services Marketing [online], 25

(5), 371-380.

The Express Tribune, 2017. Is Tesco coming to Pakistan? [online]. Karachi: The Express Tribune.

Available from: https://tribune.com.pk/story/1326865/tesco-partners-alpha-supermarket/


The Telegraph, 2016. The best Christmas 2016 adverts [online]. London: Telegraph Media

Group Limited. Available from: http://www.telegraph.co.uk/business/0/best-christmas-

adverts-of-2016/ (accessed 20 February 2017).

The World Bank, 2016. Rural population [online]. Washington: The World Bank. Available from:

https://data.worldbank.org/indicator/SP.RUR.TOTL?view=chart [Accessed 11 January 2018].

The World Factbook, 2018a. United Kingdom [online]. Washington: Central Intelligence

Agency. Available from: https://www.cia.gov/library/publications/the-world-

factbook/geos/uk.html [Accessed 11 January 2018].

The World Factbook, 2018b. Sri Lanka [online]. Washington: Central Intelligence Agency.

Available from: https://www.cia.gov/library/publications/the-world-factbook/geos/ce.html


218

Thomas, T. and Carraher, C.E., 2014. A Retail Perspective on the Shopping Behavior, Cultures,

and Personalities for China, United Arab Emirates, Belgium, India, Germany and America.

Journal of Technology Management in China [online], 9 (3), 289-296.

Triandis, H.C., 1989. The Self and Social Behavior in Differing Cultural Contexts. Psychological

Review [online], 96 (3), 506-520.

Trompenaars, F., 1993. Riding the Waves of Culture: Understanding Cultural Diversity in

Business. London: Economists Books.

Trompenaars, F. and Hampden-Turner, C., 1997. Riding the Waves of Culture: Understanding

Cultural Diversity in Business. London: Nicholas Brealey.

Tse, D.K. and Wilton, P.C., 1988. Models of Consumer Satisfaction Formation: An Extensive.

Journal of Marketing Research [online], 25 (2), 204-212.

Tucker, W.T., 1964. The Development of Brand Loyalty. Journal of Marketing Research (JMR)

[online], 1 (3), 32-35.

Uncles, M.D., Dowling, G.R. and Hammond, K., 2003. Customer Loyalty and Customer Loyalty

Programs. Journal of Consumer Marketing [online], 20 (4), 294-316.

Usunier, J.C. and Lee, J.A. 2013. Marketing Across Cultures. 6th edition. Harlow: Pearson

Education Inc.

Valenzuela, F., 2014. Switching barriers' influences on service recovery evaluation in the retail

banking industry: Construct development and testing. Australasian Marketing Journal [online],

22 (4), 296-306.

van Teijlingen, E. and Pitchforth, E., 2006. Focus Group Research in Family Planning and

Reproductive Health Care. The Journal Of Family Planning and Reproductive Health Care /

Faculty Of Family Planning & Reproductive Health Care, Royal College Of Obstetricians &

Gynaecologists [online], 32 (1), 30-32.

Venaik, S. and Brewer, P.A., 2008. Contradictions in National Culture: Hofstede vs GLOBE

[online]. In: Cantwell, J. and Kiyak, T., eds. Proceedings of the 50th Annual Meeting of the

Academy of International Business. 50th Annual Meeting of the Academy of International

Business (AIB), Milan, Italy, 30 June-3 July 2008. 274.

219

Venaik, S. and Brewer, P.A., 2010. GLOBE practices and values: A case of diminishing marginal

utility? Journal of International Business Studies [online], 41 (8), 1316-1324.

Venaik, S. and Brewer, P.A., 2016. National culture dimensions: The perpetuation of cultural

ignorance. Management Learning [online], 47 (5), 563-589.

Vesel, P. and Zabkar, V., 2010. Comprehension of Relationship Quality in the Retail

Environment. Managing Service Quality [online], 20 (3), 213-235.

Wang, W., 2008. The Interrelationship of Retailer’s Relationship Efforts and Consumers’

Attitude and Behavior. Measuring Business Excellence [online], 12, 13-28.

Williams, R., 1976. Keywords: A Vocabulary of Culture and Society. London: Fontana/Croom

Helm.

Williams, D.E., 1992. Motives for Retailer Internationalisation: Their Impact, Structure and

Implications. Journal of Marketing Management [online], 8 (1), 269-285.

Wilson, D.T., 1995. An Integrated Model of Buyer-Seller Relationships. Journal of the Academy

of Marketing Science [online], 23 (4), 335-345.

Wong, A. and Sohal, A., 2002. An Examination of the Relationship between Trust, Commitment

and Relationship Quality. International Journal of Retail and Distribution Management [online],

30, 34-50.

World Medical Association, 2004. World medical association declaration of Helsinki, Ferney-

Voltaire: World Medical Association.

Xia, L. and Kukar-Kinney, M., 2014. For our valued customers only: Examining consumer

responses to preferential treatment practices. Journal of Business Research [online], 67 (11),

2368-2375.

Yammarino, F.J., Skinner, S.J. and Childers, T.L., 1991. Understanding Mail Survey Response

Behaviour: A Meta Analysis. Public Opinion Quarterly [online], 55 (4), 613-639.

Yap, B.W., Ramayah, T. and Shahidan, W.N.W., 2012. Satisfaction and Trust on Customer

Loyalty: A PLS Approach. Business Strategy Series [online], 13 (4), 154-167.

Yeganeh, H., Zhan, S. and Sauers, D., 2009. The Applicability of Widely Employed Frameworks

in Cross-Cultural Management Research. Journal of Academic Research in Economics [online],

1, 1-24.

220

Yi, Y. and Jeon, H., 2003. Effects of Loyalty Programs on Value Perception, Program Loyalty,

and Brand Loyalty. Journal of the Academy of Marketing Science [online], 31 (3), 229-240.

Yi, Y., Jeon, H. and Choi, B., 2013. Segregation versus Aggregation in the Loyalty Program: The

Role of Perceived Uncertainty. European Journal of Marketing [online], 47 (8), 1238-1255.

Yoo, B., Donthu, N. and Lenartowicz, T., 2011. Measuring Hofstede’s Five Dimensions of

Cultural Values at the Individual Level: Development and Validation of CVSCALE. Journal of

International Consumer Marketing [online], 23 (3/4), 193-210.

Zeithaml, V.A., Berry, L.L. and Parasuraman, A., 1996. The Behavioral Consequences of Service

Quality. Journal of Marketing [online], 60 (2), 31-46.

Zhang, J., Beatty, S.E. and Walsh, G., 2008. Review and Future Directions of Cross-Cultural

Consumer Services Research. Journal of Business Research [online], 61 (3), 211-224.

Zhang, Z.J., Krishna, A. and Dhar, S.K., 2000. The Optimal Choice of Promotional Vehicles:

Front-Loaded or Rear-Loaded Incentives? Management Science [online], 46 (3), 348-362.

221

Appendix A: Focus Group Discussion Guide

Timing Action Notes

2 minutes

Thank you very much for taking part in this discussion. These focus groups are carried out as part of a research study focusing on customer shopping behaviour. This discussion will last approximately 75-90 minutes. We will first have a discussion about your shopping habits and where you normally shop. Then we will look at why you shop where you do and what makes you do so. Finally, we will try and identify through discussion any triggers that could make you shop more or less at a particular shop. We can have a casual chat at the end of this discussion if any of you would like to and you may get in touch with me should you require any information. I would like to record our discussion as I will require a recording for analysis purposes. No one will be identified and I assure you anonymity. Name will assist me in taking notes and making sure that everyone gets a chance to share their ideas. I trust that everyone will be OK with this. While everyone will get a chance to share their comments, please feel free to add to others’ comments if you have any. This is not a Q&A session but more of a chat... There is no need to follow a pattern and take turns, speak when you have something to share and I am sure we will have a good discussion. There are Post-it notes and pens for your use. Please use them if you need to write down any ideas during the discussion so that you will not forget anything. Please note that there are no right or wrong answers and all opinions are valid and valued. Thank you.

Introducing the research study to the participants Explaining the discussion sequence No one will be identified after the discussion is concluded and recording is done simply to ensure accurate transcription of the data To remind the participants that this is an open, relaxed and non-judgemental setting

1 minute

Before we start our discussion, please introduce yourself to the group. Participants can get to know one another and relax

222

Make note of seating arrangement and identify participants

5 minutes

To start off, please take two minutes to think of an item you bought very recently from a store. Write it down and share with the group when you are ready. Tell us a little bit about it...

Writing Task 1 This will be the opening question to get everyone talking and will not be used in the analysis The aim is to get them to share what they purchased, the reason for purchase and where it was purchased (It is anticipated that someone will mention a product purchased at a grocery retailer)

8 minutes

Building on the responses to the opening question: X mentioned that he/she bought Y from Z grocery store. Was it part of a larger basket or a single purchase? Is that how you normally shop for groceries? Why is that? Could the others also please share how you normally do your grocery shopping?

The first introductory question will be used to explore their general food buying habits- How they shop, where and why

8

minutes

Based on responses to the first introductory question: So is there a specific store where you normally do your grocery shopping or do you shop anywhere? Is there a particular reason/set of reasons for this?

The second introductory question that will look at where they shop and why Note down retailers and reasons

8

minutes

I see a mix of grocery retailers... Have you shopped or considered shopping at any of these retailers other than the one(s) you have mentioned? Please explain...

The first transition question. Identify and explore perceptions of grocery retailers other than their primary retailer(s). Identifying and exploring how customers compare stores

2

Thinking of where you normally shop for groceries, please take a couple of minutes and write down the main factors that influence your purchasing. Share these with the group. Why are these factors important to you?

Writing Task 2 The second transition question. Identify and explore key factors that influence their

223

minutes

purchase and grocery retailer selection decisions Moderator listens, facilitates (pausing and probing where necessary) and makes a list

8 minutes

I have made a list of the factors you have mentioned. Have we missed anything? Can we think of any other factors that could be of importance? Introduce the following if not identified;

Trust Customer service Other store image attributes identified through the literature review

Would it be possible for us to collectively discuss and rank these factors in terms of their level of importance? Are we in agreement with the ranking of these factors in terms of their importance?

The third transition question to encourage further discussion and identify any additional factors Identify factors most important to participants and encourage debate over these factors Attempting to obtain consensus through discussion and debate

6 minutes

Given where each of you normally shop, is this your favourite/preferred store; do you always/mostly shop here? Why is it that you always/mostly tend to shop there?

Key question 1. Moderator listens, pauses and probes as necessary to facilitate an in-depth discussion This may allow the researcher to identify reasons customers give for being loyal

6

minutes

Do you think that as a customer, you need to always/mostly shop at one particular grocery store? Please explain...

Key question 2. Moderator listens, pauses and probes as necessary to facilitate an in-depth discussion Identify if customers think they should or should not be loyal and the reasons they give

8

minutes

Based on previous responses: X, you mentioned loyalty programmes... What is your general opinion of these programmes? Could the others also share what they think about these programmes?

Key question 3. Identifying customer perceptions of loyalty programmes and identify possible comparisons customers may make

224

Use loyalty programme attributes identified from the literature review for probing… If needed, use cultural dimensions…

6

minutes

Do you think that you would shop more at a specific store? Can that store do anything about it?

Key question 4. Moderator listens, pauses and probes as necessary to facilitate an in-depth discussion It is anticipated that loyalty programmes will be mentioned

6 minutes

Option 1: X, you mentioned that a store can make you shop there more by using a loyalty programme... How do you think that could be done? Option 2: No one seemed to talk about loyalty programmes. Do you not think that a store can use such a programme in its efforts to make you shop more in that store? Use loyalty programme attributes identified from the literature review for probing…

Key question 5. Moderator listens, pauses and probes as necessary to facilitate an in-depth discussion Aim is to identify if they would be willing to shop more when a programme is in effect and if so, what they would expect from such a programme

5 minutes

That was a very good discussion. This is how I would summarise what we have discussed. How well do I cover it? Have I missed anything?

The summary question to ensure that the participants agree with the summary provided

5 minutes

Given what we have discussed, have we missed anything important? The insurance question to ensure that critical aspects have not been missed

1

minute

That brings us to the end of our discussion. I enjoyed listening to you and have certainly learnt something new. I thank you for your support, patience and valuable contribution to my study. Before you leave, please read and sign the form in front of you to confirm that you are in agreement to have this discussion recorded and for the data to be used in my research work.

Conclusion and getting all participants to sign the consent form

225

Appendix B: Participant Information Sheet

The title of the research project

The Impact of National Culture and Industry Structure on Grocery Retail Customer Loyalty: A

Comparative Study of the United Kingdom and Sri Lanka.

You are being invited to take part in a research project. Before you decide it is important for

you to understand why the research is being done and what it will involve. Please take time to

read the following information carefully and discuss it with others if you wish. Ask us if there is

anything that is not clear or if you would like more information. Take time to decide whether

or not you wish to take part.

What is the purpose of the project? Culture has been found to affect the way customers in different cultural settings behave.

Structure which in a nutshell is the setup of an industry also affects customer behaviour.

Despite this, little research has been to identify how these two aspects affect customer

behaviour in terms of customer loyalty in the grocery retail sector. Loyalty in grocery retailing

in general is the repetitive patronage paid to one outlet over others and a better opinion about

this outlet compared to others. This could be one standalone store or a supermarket chain.

Why have I been chosen? You have been chosen because the researcher believes that you will provide valuable insights

during focus group discussion. You also meet the criteria set for selecting potential

participants.

Do I have to take part? It is up to you to decide whether or not to take part. If you do decide to take part you will be

given this information sheet to keep (and be asked to sign a consent form) and you can still

withdraw up to the point of data being anonymised (Responses will not be attached to any

person). You do not have to give a reason.

What do I have to do? You will be required to take part in a focus group just once which will not exceed one and half

hours in duration. In the focus group, the researcher will ask several questions to which all

participants can provide answers. You should be respectful of other participants and the

researcher will ensure that all participants get an equal chance to share their ideas. You may

also contribute to ideas of others and share any differing views. There are no right or wrong

answers and all opinions are considered valuable.

226

The focus groups will be conducted by the researcher who is a PhD candidate at Bournemouth

University and the project is supervised by two senior lecturers.

What are the possible disadvantages and risks of taking part? No disadvantages or risks can be foreseen. Only your time and valuable opinions are required.

What are the possible benefits of taking part? Even though you may not benefit directly, one outcome of this research is help grocery

retailers be more effective in designing loyalty programmes. This in turn should benefit

customers like you.

Will my taking part in this project be kept confidential?/ What will happen to the results of the research project? All the information that we collect about you during the course of the research will be kept

strictly confidential. You will not be able to be identified in any reports or publications.

What type of information will be sought from me and why is the collection of this information relevant for achieving the research project’s objectives? Your opinions and ideas will be gathered and these will be used to develop theory. One output

will be a general idea of how customers from different cultural and structural background

perceive loyalty and loyalty programmes. The other is where such information will help design

a questionnaire to be used in a subsequent survey.

Contact for further information You may contact the researcher via email ([email protected]) should

your require any additional information.

Will I be recorded, and how will the recorded media be used? The audio recordings of discussion will be used only for transcribing and analysis purposes. No

other use will be made of them without your written permission and no one outside the

project will be allowed access to the original recordings.

You will be given a copy of the information sheet and a signed consent form to keep should

you decide to take part in the research.

Thank you very much for taking the time to read through this sheet and I hope you will be

willing to take part in the study.

227

Appendix C: Consent Form

Full title of project: The Impact of National Culture and Industry Structure on Grocery Retail

Customer Loyalty: A Comparative Study of the United Kingdom and Sri Lanka.

Name and Position of Researcher: Maheshan De Silva Kanakaratne- PhD Researcher Contact Details of Researcher: [email protected] Name and Position of Supervisors: Dr Jeffery Bray (Senior Lecturer) & Dr Charles McIntyre (Senior Lecturer) Contact Details of Supervisors: Please email researcher for contact details

Please Initial Here

I confirm that I have read and understood the participant information sheet for the above research project and have had the opportunity to ask questions.

I understand that my participation is voluntary and that I am free to withdraw up to the point where the data is anonymised without giving reason and without there being any negative consequences. In addition, should I not wish to answer any particular question(s), I am free to decline.

I give permission for members of the research team to have access to my anonymised responses. I understand that my name will not be linked with the research materials and I will not be identified or identifiable in the report or reports that result from the research.

I agree to take part in the above research project.

____________________________ _______________ __________________________________

Name of Participant Date Signature

Maheshan De Silva Kanakaratne

____________________________ _______________ __________________________________ Name of Researcher Date Signature

228

Appendix D: Questionnaire

232

Appendix E: Store Image Items

Item Source(s) Price Lessig (1973); Doyle and Fenwick (1974); Bearden (1977); Ghosh (1990); Thomas

(2013) Deals and offers Lindquist (1974); Ghosh (1990) Freshness Lessig (1973); Bearden (1977); Thomas (2013) Quality Lessig (1973); Bearden (1977); Thomas (2013) Range and choice Lessig (1973); Lindquist (1974); Doyle and Fenwick (1974); Bearden (1977); Ghosh

(1990); Thomas (2013) Convenience Chang and Tu (2005) Own label products

Not adapted from the literature but mentioned by participants in both countries

Parking Bearden (1977); Ghosh (1990) Ethical considerations

Not adapted from the literature but mentioned by UK participants

Loyalty Programmes

Not adapted from the literature but mentioned by participants in both countries

Customer service Lindquist (1974); Bearden (1977); Ghosh (1990); Chang and Tu (2005); Thomas (2013)

Reputation Lessig (1973); Lindquist (1974) Overall satisfaction Lindquist (1974) Distance to home/work

Not adapted from the literature but was included as a sub-element of convenience

Accessibility Doyle and Fenwick (1974); Bearden (1977); Ghosh (1990) Opening times Not adapted from the literature but mentioned by Sri Lankan participants Value for money Not adapted from the literature but mentioned by Sri Lankan participants Affiliated credit cards

Not adapted from the literature but mentioned by Sri Lankan participants

Payment options Not adapted from the literature but mentioned by Sri Lankan participants Loyalty card offers Not adapted from the literature but mentioned by participants in both countries Past experience Lessig (1973); Lindquist (1974) One-stop-shop Not adapted from the literature but mentioned by Sri Lankan participants Store layout Martineau (1958); Lessig (1973); Lindquist (1974) Trust Lessig (1973); Lindquist (1974)

Appendix F: True Loyalty Scales

Item Source(s) I will continue to shop at my main supermarket Omar et al. (2011) I would recommend my main supermarket to others Sirohi et al. (1998); Yap et al. (2012) I spend more than 50% of my grocery budget here Guenzi et al. (2009) I buy more than 50% of my groceries (quantity) here Bridson et al. (2008) I intend to use this supermarket more in the future Sirohi et al. (1998); Yap et al. (2012) I prefer my main supermarket over others Yap et al. (2012) I consider my main supermarket my first choice Omar et al. (2010); Omar et al. (2013) I have the strongest preference for this supermarket Omar et al. (2010); Omar et al. (2011) My main supermarket is my favourite supermarket Author I have no preference for my main supermarket Author

233

Appendix G: Cover Letter

234

Appendix H: Perceptual Map

235

Appendix I: EFA Pattern Matrix- UK

Variable F1 F2 F3 F4 F5 F6 F7 F8 F9 F 10 CL8 .936 CL7 .922 CL6 .909 CL9 .867 CL2 .696 CL5 .677 CL1 .655 CL4 .574 PT3 1.026 PT2 .981 PT1 .964 PT4 .824 PT5 .809 SL_LL4 .993 SL_LL2 .922 SL_LL1 .868 SL_LL3 .836 CAP4 .923 CAP5 .922 CAP3 .672 CAP1 .609 CAP6 .563 SB4 1.026 SB5 .964 SB3 .596 SB1 .536 TIE2 1.016 TIE1 .973 TIE3 .945 OL1 .866 OL2 .776 OL3 .749 OL4 .593 OL6 .430 CS3 1.003 CS4 .809 CS2 .757 CS5 .520 FP5 .774 FP1 .671 FP4 .658 FP2 .642 FP3 .434 TRA3 1.010 TRA4 .942 TRA1 .423 TRA2 .408 Eigenvalue 8.995 7.799 3.656 2.913 2.627 2.303 2.102 1.908 1.588 1.237 % Variance 19.1% 16.6% 7.8% 6.2% 5.6% 4.9% 4.5% 4.1% 3.4% 2.6% Cronbach’s Alpha 0.927 0.967 0.943 0.860 0.914 0.980 0.815 0.853 0.806 0.848

236

Appendix J: EFA Pattern Matrix- Sri Lanka

Variable F1 F2 F3 F4 F5 F6 F7 F8 F9 F 10 F11 CL8 .995 CL7 .943 CL9 .920 CL6 .865 CL5 .689 CL4 .575 CL2 .560 CL1 .462 PT4 .951 PT3 .922 PT2 .839 PT5 .777 PT1 .726 SL_LL3 .938 SL_LL2 .922 SL_LL4 .892 SL_LL1 .494 SB4 .916 SB3 .836 SB5 .811 SB1 .391 OL6 .860 OL5 .788 OL2 .657 OL4 .471 CS3 .900 CS2 .735 CS4 .660 CS5 .519 TIE2 .955 TIE3 .875 TIE1 .775 CAP4 .843 CAP5 .803 CAP3 .514 CAP6 .488 TRA4 .899 TRA3 .857 TRA2 .437 TRA1 .363 FP1 .842 FP2 .745 FP5 .649 FP3 .405 REW4 .802 REW1 .717 REW2 .551 Eigenvalue 9.793 5.191 3.665 2.405 2.326 2.193 2.037 1.627 1.483 1.394 1.114 % Variance 20.8% 11% 7.8% 5.2% 4.9% 4.7% 4.3% 3.5% 3.2% 2.9% 2.4% Cronbach’s Alpha 0.920 0.930 0.886 0.900 0.781 0.789 0.932 0.746 0.728 0.731 0.721

237

Appendix K: Composite Reliability, Converegent Validity & Divergent Valdity- UK

CR/MaxR(H) AVE MSV FP CL PT SLLL CAP SB TIE OL CS TRA

FP 0.785/0.834 0.432 0.166 0.657

CL 0.899/0.948 0.537 0.263 -0.331 0.733

PT 0.913/0.968 0.679 0.150 0.244 0.009 0.824

SLLL 0.936/0.979 0.785 0.263 0.346 -0.513 -0.017 0.886

CAP 0.855/0.983 0.549 0.085 0.291 -0.277 0.088 0.285 0.741

SB 0.859/0.985 0.606 0.219 0.034 -0.024 0.377 0.133 0.005 0.779

TIE 0.860/0.986 0.671 0.219 0.101 -0.068 0.387 -0.006 -0.011 0.468 0.819

OL 0.811/0.987 0.465 0.166 0.408 -0.119 0.094 0.086 0.116 0.036 0.185 0.682

CS 0.849/0.989 0.594 0.029 -0.048 0.095 -0.047 -0.104 -0.081 -0.075 -0.171 0.108 0.771

TRA 0.819/0.991 0.556 0.070 0.035 -0.040 -0.253 0.022 -0.026 -0.265 -0.251 0.075 0.075 0.746

Appendix L: Composite Reliability, Converegent Validity & Divergent Valdity- Sri Lanka

CR/MaxR(H) AVE MSV TRA CL PT SLLL SB OL CS TIE CAP FP

TRA 0.769/0.791 0.459 0.047 0.678

CL 0.898/0.940 0.530 0.069 -0.079 0.728

PT 0.899/0.964 0.642 0.441 -0.216 0.159 0.802

SLLL 0.873/0.973 0.639 0.055 0.045 -0.235 0.029 0.799

SB 0.871/0.980 0.635 0.441 -0.106 0.141 0.664 0.034 0.797

OL 0.716/0.981 0.464 0.035 -0.109 -0.097 -0.055 0.130 -0.065 0.681

CS 0.783/0.983 0.482 0.067 -0.027 0.259 0.225 -0.115 0.219 -0.028 0.694

TIE 0.893/0.985 0.736 0.437 -0.172 0.262 0.568 -0.047 0.661 -0.186 0.211 0.858

CAP 0.679/0.986 0.428 0.029 -0.090 -0.092 0.122 0.104 0.063 0.171 0.048 0.084 0.654

FP 0.740/0.987 0.437 0.031 0.056 -0.176 -0.078 0.059 -0.033 -0.045 -0.126 -0.047 -0.043 0.661

ã - Bournemouth Universityeprints.bournemouth.ac.uk/31168/1/DE SILVA KANAKARATNE, Mahe… · ï z...

Documents

Transcript of ã - Bournemouth Universityeprints.bournemouth.ac.uk/31168/1/DE SILVA KANAKARATNE, Mahe… · ï z...